Authors: Naimur Rahman
Abstract: Deep learning is often applied in settings where data are limited, correlated, and difficult to obtain, yet evaluation practices do not always reflect these constraints. Neuroimaging for prodromal Parkinsons disease is one such case, where subject numbers are small and individual scans produce many highly related samples. This work examines prodromal Parkinsons detection from resting-state fMRI as a machine learning problem centered on learning under extreme data scarcity. Using fMRI data from 40 subjects, including 20 prodromal Parkinsons cases and 20 healthy controls, ImageNet-pretrained convolutional neural networks are fine-tuned and evaluated under two different data partitioning strategies. Results show that commonly used image-level splits allow slices from the same subject to appear in both training and test sets, leading to severe information leakage and near-perfect accuracy. When a strict subject-level split is enforced, performance drops substantially, yielding test accuracies between 60 and 81 percent. Models with different capacity profiles are compared, including VGG19, Inception V3, Inception ResNet V2, and the lightweight MobileNet V1. Under subject-level evaluation, MobileNet demonstrates the most reliable generalization, outperforming deeper architectures despite having significantly fewer parameters. These results indicate that in extreme low-data regimes, evaluation strategy and model capacity have a greater impact on performance than architectural depth. Although the analysis is limited to a single cohort of 40 subjects and does not include external validation or cross-validation, it provides a concrete case study and practical recommendations for evaluating deep learning models under severe data scarcity.
Authors: Niklas Freund, Zekiye Ilknur-\"Oz, Tobias Klockau, Patrick Naumann, Philipp Neumaier, Martin K\"oppel
Abstract: The monitoring of the route and track environment plays an important role in automated driving. For example, it can be used as an assistance system for route monitoring in automation level Grade of Automation (GoA) 2, where the train driver is still on board. In fully automated, driverless driving at automation level GoA4, these systems finally take over environment monitoring completely independently. With the help of artificial intelligence (AI), they react automatically to risks and dangerous events on the route. To train such AI algorithms, large amounts of training data are required, which must meet high-quality standards due to their safety relevance. In this publication we present an automatic method for assuring the quality of training data, significantly reducing the manual workload and accelerating the development of these systems. We propose an open-source tool designed to detect nine common errors found in multi-sensor datasets for railway vehicles. To evaluate the performance of the framework, all detected errors were manually validated. Six issue detection methods achieved 100% precision, while three additional methods reached precision rates 96% and 97%.
Authors: Takumi Hachimine, Yuhwan Kwon, Cheng-Yu Kuo, Tomoya Yamanokuchi, Takamitsu Matsubara
Abstract: Non-destructive extraction of the target internal part, such as batteries and motors, by cutting surrounding structures is crucial at recycling and disposal sites. However, the diversity of products and the lack of information on disassembly procedures make it challenging to decide where to cut. This study explores a method for non-destructive extraction of a target internal part that iteratively estimates the internal structure from observed cutting surfaces and formulates cutting plans based on the estimation results. A key requirement is to estimate the probability of the target part's presence from partial observations. However, learning conditional generative models for this task is challenging: The high dimensionality of 3D shape representations makes learning difficult, and conventional models (e.g., conditional variational autoencoders) often fail to capture multi-modal predictive uncertainty due to mode collapse, resulting in overconfident predictions. To address these issues, we propose VoxelDiffusionCut, which iteratively estimates the internal structure represented as voxels using a diffusion model and plans cuts for non-destructive extraction of the target internal part based on the estimation results. Voxel representation allows the model to predict only attributes at fixed grid positions, i.e., types of constituent parts, making learning more tractable. The diffusion model completes the voxel representation conditioned on observed cutting surfaces, capturing uncertainty in unobserved regions to avoid erroneous cuts. Experimental results in simulation suggest that the proposed method can estimate internal structures from observed cutting surfaces and enable non-destructive extraction of the target internal part by leveraging the estimated uncertainty.
Authors: Sushi Rao, Jingwei Li
Abstract: This paper introduces a lightweight image super-resolution (SR) network, termed the Multi-scale Spatial Adaptive Attention Network (MSAAN), to address the common dilemma between high reconstruction fidelity and low model complexity in existing SR methods. The core of our approach is a novel Multi-scale Spatial Adaptive Attention Module (MSAA), designed to jointly model fine-grained local details and long-range contextual dependencies. The MSAA comprises two synergistic components: a Global Feature Modulation Module (GFM) that learns coherent texture structures through differential feature extraction, and a Multi-scale Feature Aggregation Module (MFA) that adaptively fuses features from local to global scales using pyramidal processing. To further enhance the network's capability, we propose a Local Enhancement Block (LEB) to strengthen local geometric perception and a Feature Interactive Gated Feed-Forward Module (FIGFF) to improve nonlinear representation while reducing channel redundancy. Extensive experiments on standard benchmarks (Set5, Set14, B100, Urban100, Manga109) across $\times2$, $\times3$, and $\times4$ scaling factors demonstrate that both our lightweight (MSAAN-light) and standard (MSAAN) versions achieve superior or competitive performance in terms of PSNR and SSIM, while maintaining significantly lower parameters and computational costs than state-of-the-art methods. Ablation studies validate the contribution of each component, and visual results show that MSAAN reconstructs sharper edges and more realistic textures.
Authors: M Iffat Hossain, Laura Brattain
Abstract: During image-guided procedures, real-time image segmentation is often required. This demands lightweight AI models that can operate on resource-constrained devices. One important use case is endoscopy-guided colonoscopy, where polyps must be detected in real time. The Kvasir-Seg dataset, a publicly available benchmark for this task, contains 1,000 high-resolution endoscopic images of polyps with corresponding pixel-level segmentation masks. Achieving real-time inference speed for clinical deployment in constrained environments requires highly efficient and lightweight network architectures. However, many existing models remain too computationally intensive for embedded deployment. Lightweight architectures, although faster, often suffer from reduced spatial precision and weaker contextual understanding, leading to degraded boundary quality and reduced diagnostic reliability. To address these challenges, we introduce BiSe-UNet, a lightweight dual-path U-Net that integrates an attention-refined context path with a shallow spatial path for detailed feature preservation, followed by a depthwise separable decoder for efficient reconstruction. Evaluated on the Kvasir-Seg dataset, BiSe-UNet achieves competitive Dice and IoU scores while sustaining real-time throughput exceeding 30 FPS on Raspberry Pi 5, demonstrating its effectiveness for accurate, lightweight, and deployable medical image segmentation on edge hardware.
Authors: Aman Ulla
Abstract: Document extraction is an important step before retrieval-augmented generation (RAG), knowledge bases, and downstream generative AI can work. It turns unstructured documents like PDFs and scans into structured text and layout-aware representations. We introduce NovaLAD, a comprehensive document parsing system that integrates two concurrent YOLO object detection models - element detection and layout detection - with rule-based grouping and optional vision-language enhancement. When a page image is sent in, the first thing that happens is that it goes through both models at the same time. The element model finds semantic content like the title, header, text, table, image, and so on, and the layout model finds structural regions like layout_box, column_group, multi_column, row_group, and so on. A key design decision is to first send an image or figure through an image classifier (ViT) that decides whether it is relevant or not. Only useful images are then submitted to the Vision LLM for title, summary, and structured information, which cuts down on noise and costs. NovaLAD is built for speed: it works on CPU, employs parallel execution for detection, classification, OCR, and conversion, and generates several forms, including structured JSON, Markdown, RAG-ready texts, and knowledge graphs. We test on the DP-Bench benchmark (upstage/dp-bench) and get 96.49% TEDS and 98.51% NID, which is better than both commercial and open-source parsers. This paper explains how to extract data, how the architecture works, how data flows, and how to make NovaLAD both accurate and usable without needing a GPU.
Authors: Yannian Gu, Xizhuo Zhang, Linjie Mu, Yongrui Yu, Zhongzhen Huang, Shaoting Zhang, Xiaofan Zhang
Abstract: Recent advances in Large Vision-Language Models (LVLMs) have shown strong potential for multi-modal radiological reasoning, particularly in tasks like diagnostic visual question answering (VQA) and radiology report generation. However, most existing approaches for 3D CT analysis largely rely on static, single-pass inference. In practice, clinical interpretation is a dynamic, tool-mediated workflow where radiologists iteratively review slices and use measurement, radiomics, and segmentation tools to refine findings. To bridge this gap, we propose CT-Flow, an agentic framework designed for interoperable volumetric interpretation. By leveraging the Model Context Protocol (MCP), CT-Flow shifts from closed-box inference to an open, tool-aware paradigm. We curate CT-FlowBench, the first large-scale instruction-tuning benchmark tailored for 3D CT tool-use and multi-step reasoning. Built upon this, CT-Flow functions as a clinical orchestrator capable of decomposing complex natural language queries into automated tool-use sequences. Experimental evaluations on CT-FlowBench and standard 3D VQA datasets demonstrate that CT-Flow achieves state-of-the-art performance, surpassing baseline models by 41% in diagnostic accuracy and achieving a 95% success rate in autonomous tool invocation. This work provides a scalable foundation for integrating autonomous, agentic intelligence into real-world clinical radiology.
Authors: Edouard Lansiaux, Margaux Leman, Mehdi Ammi
Abstract: Clear aligner therapy now dominates orthodontics, yet clinician review of digitally planned tooth movements-typically via ClinCheck (Align Technology)-remains slow and error-prone. We present OrthoAI, an open-source proof-of-concept decision-support system combining lightweight 3D dental segmentation with automated biomechanical analysis to assist treatment-plan evaluation. The framework uses a Dynamic Graph CNN trained on landmark-reconstructed point clouds from 3DTeethLand (MICCAI) and integrates a rule-based biomechanical engine grounded in orthodontic evidence (Kravitz et al 2009; Simon et al 2014). The system decomposes per-tooth motion across six degrees of freedom, computes movement-specific predictability, issues alerts when biomechanical limits are exceeded, and derives an exploratory composite index. With 60,705 trainable parameters, segmentation reaches a Tooth Identification Rate of $81.4\%$ and mIoU of $8.25\%$ on surrogate point clouds-reflecting sparse landmark supervision rather than dense meshes. Although spatial boundaries are coarse, downstream analysis depends mainly on tooth identity and approximate centroid/axis estimation. Results establish a baseline for future full-mesh training and highlight current perceptual limits. The end-to-end pipeline runs in $<4s$ on consumer hardware. Code, weights, and analysis tools are released to support reproducible research in geometric deep learning and digital orthodontics. The system has not been validated on real intraoral meshes and should not be assumed to generalize beyond landmark-derived representations.
Authors: Miao Zhang, Ruixiao Zhang, Jianxin Shi, Hengzhi Wang, Hao Fang, Jiangchuan Liu
Abstract: Video-language models (VLMs) are reshaping video querying services, bringing unified solutions to complex perception and reasoning tasks. However, deploying large VLMs in real-world systems remains challenging due to their high resource demands, and remote-based deployment often results in unacceptable response delays. Although small, locally deployable VLMs offer faster responses, they unavoidably fall short in accuracy. To reconcile this trade-off, we propose QuickGrasp, a responsive, quality of service (QoS)-aware system that bridges this gap through a local-first architecture with on-demand edge augmentation. Built upon the highly modular architecture of VLMs, QuickGrasp shares the vision representation across model variants to avoid redundant computation. To maximize system-wide efficiency, QuickGrasp introduces three key designs: accelerated video tokenization, query-adaptive edge augmentation, and delay-aware, accuracy-preserving vision token density configuration. We implement a prototype of QuickGrasp and evaluate it across multiple video understanding benchmarks. The results show that QuickGrasp matches the accuracy of large VLMs while achieving up to a 12.8x reduction in response delay. QuickGrasp represents a key advancement toward building responsive video querying services for open-world understanding that fully leverage the capabilities of VLMs.
Authors: Kaustav Das, Gaston Rauchs, Jan Sykora, Anna Kucerova
Abstract: This work tests a self-annotation-based unsupervised methodology for training a convolutional neural network (CNN) model for semantic segmentation of X-ray computed tomography (XCT) scans of concretes. Concrete poses a unique challenge for XCT imaging due to similar X-ray attenuation coefficients of aggregates and mortar, resulting in low-contrast between the two phases in the ensuing images. While CNN-based models are a proven technique for semantic segmentation in such challenging cases, they typically require labeled training data, which is often unavailable for new datasets or are costly to obtain. To counter that limitation, a self-annotation technique is used here which leverages superpixel algorithms to identify perceptually similar local regions in an image and relates them to the global context in the image by utilizing the receptive field of a CNN-based model. This enables the model to learn a global-local relationship in the images and enables identification of semantically similar structures. We therefore present the performance of the unsupervised training methodology on our XCT datasets and discuss potential avenues for further improvements.
Authors: Hugo Majer, Martin Fleischmann
Abstract: The Local Climate Zone (LCZ) framework is commonly employed to represent urban form in morphological analyses despite its mapping predominantly relies on satellite imagery. Urban morphometrics, describing urban form via numerical measures of physical aspects and spatial relationships of its elements, offers another avenue. This study evaluates the ability of morphometric assessment to predict LCZs using a) a morphometric-based LCZ prediction, and b) a fusion-based LCZ prediction combining morphometrics with satellite imagery. We calculate 321 2D morphometric attributes from building footprints and street networks, covering their various properties at multiple spatial scales. Subsequently, we develop four classification schemes: morphometric-based prediction, baseline image-based prediction, and two techniques fusing morphometrics with imagery. We evaluate them across five sites. Results from the morphometric-based prediction indicate that the correspondence between 2D urban morphometrics and urban LCZ types is selective and inconsistent, rendering the efficacy of this method site-dependent. Nevertheless, it demonstrated that a much broader range of urban form properties is relevant for distinguishing LCZ types compared to standard parameters. Relative to the image-based baseline, the fusion yielded relatively distinct accuracy improvements for urban LCZ types at two sites; however, gains at the remaining sites were negligible or even slightly negative, suggesting that the benefits of fusion are modest and inconsistent. Collectively, these results indicate that the relationship between the LCZs and the measurable, visible aspects of urban form is tenuous, thus the LCZ framework should be used with caution in morphological studies.
Authors: Kairan Zhao, Eleni Triantafillou, Peter Triantafillou
Abstract: Generative models have been shown to "memorize" certain training data, leading to verbatim or near-verbatim generating images, which may cause privacy concerns or copyright infringement. We introduce Guidance Using Attractive-Repulsive Dynamics (GUARD), a novel framework for memorization mitigation in text-to-image diffusion models. GUARD adjusts the image denoising process to guide the generation away from an original training image and towards one that is distinct from training data while remaining aligned with the prompt, guarding against reproducing training data, without hurting image generation quality. We propose a concrete instantiation of this framework, where the positive target that we steer towards is given by a novel method for (cross) attention attenuation based on (i) a novel statistical mechanism that automatically identifies the prompt positions where cross attention must be attenuated and (ii) attenuating cross-attention in these per-prompt locations. The resulting GUARD offers a surgical, dynamic per-prompt inference-time approach that, we find, is by far the most robust method in terms of consistently producing state-of-the-art results for memorization mitigation across two architectures and for both verbatim and template memorization, while also improving upon or yielding comparable results in terms of image quality.
Authors: Bibin Wilson
Abstract: Zero-shot object detection enables recognising novel objects without task-specific training, but current approaches rely on large vision language models (VLMs) like CLIP that require hundreds of megabytes of memory - far exceeding the constraints of micro controller units (MCUs). We present TinyVLM, the first framework enabling zero-shot object detection on resource-constrained MCUs with less than 1MB of memory. Our approach introduces three key innovations: (1) a decoupled architecture that separates visual inference from text encoding, allowing precomputed class embeddings to be stored in flash memory; (2) Matryoshka distillation that trains nested embeddings at multiple dimensions (16-256), enabling flexible accuracy-memory trade-offs; and (3) quantized embedding storage that reduces class prototype memory by 4x with minimal accuracy loss. Trained on Conceptual Captions 3M (CC3M), TinyVLM achieves competitive zero-shot accuracy on COCO, Flowers102, and Food101 while requiring only 285KB of RAM and 892KB of flash memory for the deployed vision encoder. We demonstrate real-time inference at 26 FPS on STM32H7 and over 1,000 FPS on MAX78000 with its CNN accelerator, enabling practical zero-shot detection on edge devices for the first time.
Authors: Bibin Wilson
Abstract: Deploying object detection on microcontrollers (MCUs) enables intelligent edge devices but current models cannot learn new object categories after deployment. Existing continual learning methods require storing raw images far exceeding MCU memory budgets of tens of kilobytes. We present Latent Replay Detection (LRD), the first framework for continual object detection under MCU memory constraints. Our key contributions are: 1. Task-Adaptive Compression: Unlike fixed PCA, we propose learnable compression with FiLM (Feature-wise Linear Modulation) conditioning, where task specific embeddings modulate the compression to preserve discriminative features for each task's distribution; 2. Spatial-Diverse Exemplar Selection: Traditional sampling ignores spatial information critical for detection - we select exemplars maximizing bounding box diversity via farthest-point sampling in IoU space, preventing localization bias in replay; 3. MCU-Deployable System: Our latent replay stores 150 bytes per sample versus >10KB for images, enabling a 64KB buffer to hold 400+ exemplars. Experiments on CORe50 (50 classes, 5 tasks) demonstrate that LRD achieves mAP@50 on the initial task and maintains strong performance across subsequent tasks - a significant improvement over naive fine-tuning while operating within strict MCU constraints. Our task-adaptive FiLM compression and spatial diverse exemplar selection work synergistically to preserve detection capabilities. Deployed on STM32H753ZI, ESP32-S3, and MAX78000 MCUs, LRD achieves 4.9-97.5ms latency per inference within a 64KB memory budget-enabling practical continual detection on edge devices for the first time.
Authors: Andreas Tritsarolis, Toma\v{z} Bokan, Matej Brumen, Domen Mongus, Yannis Theodoridis
Abstract: The modernization of agriculture has motivated the development of advanced analytics and decision-support systems to improve resource utilization and reduce environmental impacts. Targeted Spraying and Fertilization (TSF) is a critical operation that enables farmers to apply inputs more precisely, optimizing resource use and promoting environmental sustainability. However, accurate TSF is a challenging problem, due to external factors such as crop type, fertilization phase, soil conditions, and weather dynamics. In this paper, we present TerrAI, a Neural Network-based solution for TSF, which considers the spatio-temporal variability across different parcels. Our experimental study over a real-world remote sensing dataset validates the soundness of TerrAI on data-driven agricultural practices.
Authors: Sathwik Karnik, Juyeop Kim, Sanmi Koyejo, Jong-Seok Lee, Somil Bansal
Abstract: Text-to-image diffusion models often memorize training data, revealing a fundamental failure to generalize beyond the training set. Current mitigation strategies typically sacrifice image quality or prompt alignment to reduce memorization. To address this, we propose Reachability-Aware Diffusion Steering (RADS), an inference-time framework that prevents memorization while preserving generation fidelity. RADS models the diffusion denoising process as a dynamical system and applies concepts from reachability analysis to approximate the "backward reachable tube"--the set of intermediate states that inevitably evolve into memorized samples. We then formulate mitigation as a constrained reinforcement learning (RL) problem, where a policy learns to steer the trajectory away from memorization via minimal perturbations in the caption embedding space. Empirical evaluations show that RADS achieves a superior Pareto frontier between generation diversity (SSCD), quality (FID), and alignment (CLIP) compared to state-of-the-art baselines. Crucially, RADS provides robust mitigation without modifying the diffusion backbone, offering a plug-and-play solution for safe generation. Our website is available at: https://s-karnik.github.io/rads-memorization-project-page/.
URLs: https://s-karnik.github.io/rads-memorization-project-page/.
Authors: Xiangyan Qu, Zhenlong Yuan, Jing Tang, Rui Chen, Datao Tang, Meng Yu, Lei Sun, Yancheng Bai, Xiangxiang Chu, Gaopeng Gou, Gang Xiong, Yujun Cai
Abstract: Image Chain-of-Thought (Image-CoT) is a test-time scaling paradigm that improves image generation by extending inference time. Most Image-CoT methods focus on text-to-image (T2I) generation. Unlike T2I generation, image editing is goal-directed: the solution space is constrained by the source image and instruction. This mismatch causes three challenges when applying Image-CoT to editing: inefficient resource allocation with fixed sampling budgets, unreliable early-stage verification using general MLLM scores, and redundant edited results from large-scale sampling. To address this, we propose ADaptive Edit-CoT (ADE-CoT), an on-demand test-time scaling framework to enhance editing efficiency and performance. It incorporates three key strategies: (1) a difficulty-aware resource allocation that assigns dynamic budgets based on estimated edit difficulty; (2) edit-specific verification in early pruning that uses region localization and caption consistency to select promising candidates; and (3) depth-first opportunistic stopping, guided by an instance-specific verifier, that terminates when intent-aligned results are found. Extensive experiments on three SOTA editing models (Step1X-Edit, BAGEL, FLUX.1 Kontext) across three benchmarks show that ADE-CoT achieves superior performance-efficiency trade-offs. With comparable sampling budgets, ADE-CoT obtains better performance with more than 2x speedup over Best-of-N.
Authors: Sevda \"O\u{g}\"ut, C\'edric Vincent-Cuaz, Natalia Dubljevic, Carlos Hurtado, Vaishnavi Subramanian, Pascal Frossard, Dorina Thanou
Abstract: Self-supervised vision models have achieved notable success in digital pathology. However, their domain-agnostic transformer architectures are not originally designed to account for fundamental biological elements of histopathology images, namely cells and their complex interactions. In this work, we hypothesize that a biologically-informed modeling of tissues as cell graphs offers a more efficient representation learning. Thus, we introduce GrapHist, a novel graph-based self-supervised learning framework for histopathology, which learns generalizable and structurally-informed embeddings that enable diverse downstream tasks. GrapHist integrates masked autoencoders and heterophilic graph neural networks that are explicitly designed to capture the heterogeneity of tumor microenvironments. We pre-train GrapHist on a large collection of 11 million cell graphs derived from breast tissues and evaluate its transferability across in- and out-of-domain benchmarks. Our results show that GrapHist achieves competitive performance compared to its vision-based counterparts in slide-, region-, and cell-level tasks, while requiring four times fewer parameters. It also drastically outperforms fully-supervised graph models on cancer subtyping tasks. Finally, we also release five graph-based digital pathology datasets used in our study at https://huggingface.co/ogutsevda/datasets , establishing the first large-scale graph benchmark in this field. Our code is available at https://github.com/ogutsevda/graphist .
URLs: https://huggingface.co/ogutsevda/datasets, https://github.com/ogutsevda/graphist
Authors: Zichen Geng, Zeeshan Hayder, Bo Miao, Jian Liu, Wei Liu, Ajmal Mian
Abstract: Generating realistic 3D Human-Human Interaction (HHI) requires coherent modeling of the physical plausibility of the agents and their interaction semantics. Existing methods compress all motion information into a single latent representation, limiting their ability to capture fine-grained actions and inter-agent interactions. This often leads to semantic misalignment and physically implausible artifacts, such as penetration or missed contact. We propose Disentangled Hierarchical Variational Autoencoder (DHVAE) based latent diffusion for structured and controllable HHI generation. DHVAE explicitly disentangles the global interaction context and individual motion patterns into a decoupled latent structure by employing a CoTransformer module. To mitigate implausible and physically inconsistent contacts in HHI, we incorporate contrastive learning constraints with our DHVAE to promote a more discriminative and physically plausible latent interaction space. For high-fidelity interaction synthesis, DHVAE employs a DDIM-based diffusion denoising process in the hierarchical latent space, enhanced by a skip-connected AdaLN-Transformer denoiser. Extensive evaluations show that DHVAE achieves superior motion fidelity, text alignment, and physical plausibility with greater computational efficiency.
Authors: Kangyuan Zheng, Xuan Cai, Jiangqi Wang, Guixing Fu, Zhuoshuo Li, Yazhou Chen, Xinting Ge, Liangqiong Qu, Mengting Liu
Abstract: Magnetic Resonance Imaging (MRI) is a crucial non-invasive imaging modality. In routine clinical practice, multi-stack thick-slice acquisitions are widely used to reduce scan time and motion sensitivity, particularly in challenging scenarios such as fetal brain imaging. However, the resulting severe through-plane anisotropy compromises volumetric analysis and downstream quantitative assessment, necessitating robust reconstruction of isotropic high-resolution volumes. Implicit neural representation methods, while achieving high quality, suffer from computational inefficiency due to complex network structures. We present M-Gaussian, adapting 3D Gaussian Splatting to MRI reconstruction. Our contributions include: (1) Magnetic Gaussian primitives with physics-consistent volumetric rendering, (2) neural residual field for high-frequency detail refinement, and (3) multi-resolution progressive training. Our method achieves an optimal balance between quality and speed. On the FeTA dataset, M-Gaussian achieves 40.31 dB PSNR while being 14 times faster, representing the first successful adaptation of 3D Gaussian Splatting to multi-stack MRI reconstruction.
Authors: Carlos Monroy, Benjamin Navarro
Abstract: Image segmentation and image recognition are well established computational techniques in the broader discipline of image processing. Segmentation allows to locate areas in an image, while recognition identifies specific objects within an image. These techniques have shown remarkable accuracy with modern images, mainly because the amount of training data is vast. Achieving similar accuracy in digitized images of centuries-old documents is more challenging. This difficulty is due to two main reasons: first, the lack of sufficient training data, and second, because the degree of specialization in a given domain. Despite these limitations, the ability to segment and recognize objects in these collections is important for automating the curation, cataloging, and dissemination of knowledge, making the contents of priceless collections accessible to scholars and the general public. In this paper, we report on our ongoing work in segmenting and labeling images pertaining to shipbuilding treatises from the XVI and XVII centuries, a historical period known as the Age of Exploration. To this end, we leverage SAM2 for image segmentation; Florence2 and ChatGPT for labeling; and a specialized ontology ontoShip and glossary glosShip of nautical architecture for enhancing the labeling process. Preliminary results demonstrate the potential of marrying these technologies for improving curation and retrieval of priceless historical documents. We also discuss the challenges and limitations encountered in this approach and ideas on how to overcome them in the future.
Authors: Binesh Sadanandan, Vahid Behzadan
Abstract: Medical Vision-Language Models can give different yes or no answers to rephrasings of the same clinical question. We study this in MedGemma-4B using PSF-Med Sadanandan and Behzadan (2025), which provides paraphrase pairs for systematic consistency evaluation on medical VQA. On MIMIC-CXR binary questions (n = 158), the baseline flip rate is 14.6% and mean margin difference is 1.63 logits. We validate that Gemma Scope 2 Sparse Autoencoders (SAEs) transfer to MedGemma activations, achieving R2 ~= 0.997 on both medical and general text (n = 100 prompts each, p < 0.001 for exceeding a 0.95 threshold). We then fine-tune Low-Rank Adaptation (LoRA) adapters with a combined loss that balances paraphrase consistency with answer accuracy. This combined approach prevents mode collapse that occurs with pure consistency training while reducing flip rate from 14.6% to 4.4% (p = 0.002, two-proportion z-test) and margin difference from 1.63 to 0.33 (79.5% reduction). Accuracy remains stable at 84.2% baseline versus 82.3% after training (-1.9pp, not significant). On PadChest Balanced (n = 250), flip rate drops from 13.6% to 7.8%, mean margin difference drops from 1.08 to 0.35 (67.9% reduction), and accuracy increases from 66.4% to 69.4%. A layer-range ablation shows that early layers reduce margin differences more than mechanistically selected middle layers.
Authors: Zhihao Li, Shengwei Dong, Chuang Yi, Junxuan Gao, Zhilu Lai, Zhiqiang Liu, Wei Wang, Guangtao Zhang
Abstract: Existing image SR and generic diffusion models transfer poorly to fluid SR: they are sampling-intensive, ignore physical constraints, and often yield spectral mismatch and spurious divergence. We address fluid super-resolution (SR) with \textbf{ReMD} (\underline{Re}sidual-\underline{M}ultigrid \underline{D}iffusion), a physics-consistent diffusion framework. At each reverse step, ReMD performs a \emph{multigrid residual correction}: the update direction is obtained by coupling data consistency with lightweight physics cues and then correcting the residual across scales; the multiscale hierarchy is instantiated with a \emph{multi-wavelet} basis to capture both large structures and fine vortical details. This coarse-to-fine design accelerates convergence and preserves fine structures while remaining equation-free. Across atmospheric and oceanic benchmarks, ReMD improves accuracy and spectral fidelity, reduces divergence, and reaches comparable quality with markedly fewer sampling steps than diffusion baselines. Our results show that enforcing physics consistency \emph{inside} the diffusion process via multigrid residual correction and multi-wavelet multiscale modeling is an effective route to efficient fluid SR. Our code are available on https://github.com/lizhihao2022/ReMD.
Authors: Zihang Zou, Boqing Gong, Liqiang Wang
Abstract: In this paper, we highlight a critical threat posed by emerging neural models: data plagiarism. We demonstrate how modern neural models (e.g., diffusion models) can replicate copyrighted images, even when protected by advanced watermarking techniques. To expose vulnerabilities in copyright protection and facilitate future research, we propose a general approach to neural plagiarism that can either forge replicas of copyrighted data or introduce copyright ambiguity. Our method, based on "anchors and shims", employs inverse latents as anchors and finds shim perturbations that gradually deviate the anchor latents, thereby evading watermark or copyright detection. By applying perturbations to the cross-attention mechanism at different timesteps, our approach induces varying degrees of semantic modification in copyrighted images, enabling it to bypass protections ranging from visible trademarks and signatures to invisible watermarks. Notably, our method is a purely gradient-based search that requires no additional training or fine-tuning. Experiments on MS-COCO and real-world copyrighted images show that diffusion models can replicate copyrighted images, underscoring the urgent need for countermeasures against neural plagiarism.
Authors: Haoxiang Sun, Tao Wang, Chenwei Tang, Li Yuan, Jiancheng Lv
Abstract: Following the success of Group Relative Policy Optimization (GRPO) in foundation LLMs, an increasing number of works have sought to adapt GRPO to Visual Large Language Models (VLLMs) for visual perception tasks (e.g., detection and segmentation). However, much of this line of research rests on a long-standing yet unexamined assumption: training paradigms developed for language reasoning can be transferred seamlessly to visual perception. Our experiments show that this assumption is not valid, revealing intrinsic differences between reasoning-oriented and perception-oriented settings. Using reasoning segmentation as a representative case, we surface two overlooked factors: (i) the need for a broader output space, and (ii) the importance of fine-grained, stable rewards. Building on these observations, we propose Dr.~Seg, a simple, plug-and-play GRPO-based framework consisting of a Look-to-Confirm mechanism and a Distribution-Ranked Reward module, requiring no architectural modifications and integrating seamlessly with existing GRPO-based VLLMs. Extensive experiments demonstrate that Dr.~Seg improves performance in complex visual scenarios while maintaining strong generalization. Code and models will be available at https://github.com/xVI-group-SCU/Dr-Seg.
Authors: Wenxin Tang, Jingyu Xiao, Yanpei Gong, Fengyuan Ran, Tongchuan Xia, Junliang Liu, Man Ho Lam, Wenxuan Wang, Michael R. Lyu
Abstract: Automated academic poster generation aims to distill lengthy research papers into concise, visually coherent presentations. Existing Multimodal Large Language Models (MLLMs) based approaches, however, suffer from three critical limitations: low information density in full-paper inputs, excessive token consumption, and unreliable layout verification. We present EfficientPosterGen, an end-to-end framework that addresses these challenges through semantic-aware retrieval and token-efficient multimodal generation. EfficientPosterGen introduces three core innovations: (1) Semantic-aware Key Information Retrieval (SKIR), which constructs a semantic contribution graph to model inter-segment relationships and selectively preserves important content; (2) Visual-based Context Compression (VCC), which renders selected text segments into images to shift textual information into the visual modality, significantly reducing token usage while generating poster-ready bullet points; and (3) Agentless Layout Violation Detection (ALVD), a deterministic color-gradient-based algorithm that reliably detects content overflow and spatial sparsity without auxiliary MLLMs. Extensive experiments demonstrate that EfficientPosterGen achieves substantial improvements in token efficiency and layout reliability while maintaining high poster quality, offering a scalable solution for automated academic poster generation. Our code is available at https://github.com/vinsontang1/EfficientPosterGen-Code.
URLs: https://github.com/vinsontang1/EfficientPosterGen-Code.
Authors: Saivan Talaei, Fatemeh Daneshfar, Abdulhady Abas Abdullah, Mustaqeem Khan
Abstract: Medical image segmentation is a cornerstone of computer-assisted diagnosis and treatment planning. While recent multimodal vision-language models have shown promise in enhancing semantic understanding through textual descriptions, their resilience in "in-the-wild" clinical settings-characterized by scarce annotations and hardware-induced image degradations-remains under-explored. We introduce BiCLIP (Bidirectional and Consistent Language-Image Processing), a framework engineered to bolster robustness in medical segmentation. BiCLIP features a bidirectional multimodal fusion mechanism that enables visual features to iteratively refine textual representations, ensuring superior semantic alignment. To further stabilize learning, we implement an augmentation consistency objective that regularizes intermediate representations against perturbed input views. Evaluation on the QaTa-COV19 and MosMedData+ benchmarks demonstrates that BiCLIP consistently surpasses state-of-the-art image-only and multimodal baselines. Notably, BiCLIP maintains high performance when trained on as little as 1% of labeled data and exhibits significant resistance to clinical artifacts, including motion blur and low-dose CT noise.
Authors: Bryceton Bible, Shah Md Nehal Hasnaeen, Hairong Qi
Abstract: Visibility of natural landmarks such as Mount Fuji is a defining factor in both tourism planning and visitor experience, yet it remains difficult to predict due to rapidly changing atmospheric conditions. We present FujiView, a multimodal learning framework and dataset for predicting scenic visibility by fusing webcam imagery with structured meteorological data. Our late-fusion approach combines image-derived class probabilities with numerical weather features to classify visibility into five categories. The dataset currently comprises over 100,000 webcam images paired with concurrent and forecasted weather conditions from more than 40 cameras around Mount Fuji, and continues to expand; it will be released to support further research in environmental forecasting. Experiments show that YOLO-based vision features dominate short-term horizons such as "nowcasting" and "samedaycasting", while weather-driven forecasts increasingly take over as the primary predictive signal beyond $+1$d. Late fusion consistently yields the highest overall accuracy, achieving ACC of approx 0.89 for same-day prediction and up to 84% for next-day forecasts. These results position Scenic Visibility Forecasting (SVF) as a new benchmark task for multimodal learning.
Authors: Weiting Tan, Andy T. Liu, Ming Tu, Xinghua Qu, Philipp Koehn, Lu Lu
Abstract: Generating realistic talking-head videos remains challenging due to persistent issues such as imperfect lip synchronization, unnatural motion, and evaluation metrics that correlate poorly with human perception. We propose FlowPortrait, a reinforcement-learning framework for audio-driven portrait animation built on a multimodal backbone for autoregressive audio-to-video generation. FlowPortrait introduces a human-aligned evaluation system based on Multimodal Large Language Models (MLLMs) to assess lip-sync accuracy, expressiveness, and motion quality. These signals are combined with perceptual and temporal consistency regularizers to form a stable composite reward, which is used to post-train the generator via Group Relative Policy Optimization (GRPO). Extensive experiments, including both automatic evaluations and human preference studies, demonstrate that FlowPortrait consistently produces higher-quality talking-head videos, highlighting the effectiveness of reinforcement learning for portrait animation.
Authors: Boyang Deng, Yuzhen Lu
Abstract: Developing robust models for precision vegetable weeding is currently constrained by the scarcity of large-scale, annotated weed-crop datasets. To address this limitation, this study proposes a foundational crop-weed detection model by integrating heterogeneous datasets and leveraging self-supervised learning. A total of 618,642 crop-weed images were initially collected and subsequently refined to 199,388 filtered images for fine-tuning a DINOv3 vision transformer (ViT-small) through a sequential curation strategy. The fine-tuned DINOv3 backbone was then integrated into YOLO26, serving either as a primary backbone or part of a dual-backbone architecture. A feature alignment loss was introduced in the dual backbone framework to enhance feature fusion with minimal computational overhead. Experimental results show that the proposed DINOv3-finetuned ViT-small-based YOLO26-large achieved up to a +5.4% mAP50 gain on in-domain images collected in the 2025 season. Moreover, it demonstrated strong cross-domain generalization with mAP50 improvements of +14.0% on the 2021-2023 season dataset and +11.9% on the 2024 season dataset, compared to the standard YOLO26-large. Although the DINOv3-YOLO26-large model has 45.6% more parameters and a 2.9x increase in inference latency, it maintains real-time performance at ~28.5 frames per second (fps). The curated dataset and software programs developed in this study will be made publicly available.
Authors: S. Kalaycioglu, C. Hong, M. Zhu, H. Xie
Abstract: Early ophthalmic screening in low-resource and remote settings is constrained by access to specialized equipment and trained practitioners. We present SKINOPATHY AI, a smartphone-first web application that delivers five complementary, explainable screening modules entirely through commodity mobile hardware: (1) redness quantification via LAB a* color-space normalization; (2) blink-rate estimation using MediaPipe FaceMesh Eye Aspect Ratio (EAR) with adaptive thresholding; (3) pupil light reflex characterization through Pupil-to-Iris Ratio (PIR) time-series analysis; (4) scleral color indexing foricterus and anemia proxies via LAB/HSV statistics; and (5) iris-landmark-calibrated lesion encroachment measurement with millimeter-scale estimates and longitudinal trend tracking. The system is implemented as a React/FastAPI stack with OpenCV and MediaPipe, MongoDB-backed session persistence, and PDF report generation. All algorithms are fully deterministic, privacy-preserving, and designed for non-diagnostic consumer triage. We detail system architecture, algorithm design, evaluation methodology, clinical context, and ethical boundaries of the platform. SKINOPATHY AI demonstrates that multi-signal ophthalmic screening is feasible on unmodified smartphones without cloud-based AI inference, providing a foundation for future clinically validated mobile ophthalmoscopy tools.
Authors: Nicholas Korcynski
Abstract: The binary segmentation of whiteboard strokes is hindered by extreme class imbalance, caused by stroke pixels that constitute only $1.79%$ of the image on average, and in addition, the thin-stroke subset averages $1.14% \pm 0.41%$ in the foreground. Standard region metrics (F1, IoU) can mask thin-stroke failures because the vast majority of the background dominates the score. In contrast, adding boundary-aware metrics and a thin-subset equity analysis changes how loss functions rank and exposes hidden trade-offs. We contribute an evaluation protocol that jointly examines region metrics, boundary metrics (BF1, B-IoU), a core/thin-subset equity analysis, and per-image robustness statistics (median, IQR, worst-case) under seeded, multi-run training with non-parametric significance testing. Five losses -- cross-entropy, focal, Dice, Dice+focal, and Tversky -- are trained three times each on a DeepLabV3-MobileNetV3 model and evaluated on 12 held-out images split into core and thin subsets. Overlap-based losses improve F1 by more than 20 points over cross-entropy ($0.663$ vs $0.438$, $p < 0.001$). In addition, the boundary metrics confirm that the gain extends to the precision of the contour. Adaptive thresholding and Sauvola binarization at native resolution achieve a higher mean F1 ($0.787$ for Sauvola) but with substantially worse worst-case performance (F1 $= 0.452$ vs $0.565$ for Tversky), exposing a consistency-accuracy trade-off: classical baselines lead on mean F1 while the learned model delivers higher worst-case reliability. Doubling training resolution further increases F1 by 12.7 points.
Authors: Zhaodong Wu, Haochen Xue, Qi Cao, Wenqi Mo, Yu Pei, Wenqi Xu, Jionglong Su, Yang Liu
Abstract: Thinking with Images improves fine-grained VQA for MLLMs by emphasizing visual cues. However, tool-augmented methods depend on the capacity of grounding, which remains unreliable for MLLMs. In parallel, attention-driven methods to crop the Region of Interest (ROIs) are proposed but they are constrained by (1) fragmented attention signals scattered across layers, leading to suboptimal localization and (2) relying on question- or redundant-text-conditioned attention extraction. Our analysis reveals three patterns: MLLMs may attend to the correct region yet generate incorrect coordinates, where-to-look attention is often fragmented across layers, and attention extraction is query-sensitive. Motivated by these, We propose ConFoThinking, a Consolidated-Focused-Attention-Driven Thinking framework that learns to aggregate attention into a designated intermediate layer, from which we mine and zoom in salient regions for downstream visual understanding. Moreover, we extract attention using concise semantic cues of what to look into, which mitigates the semantic noise introduced by question- or redundant-text-based attention extraction. Experiments across five VQA benchmarks demonstrate ConFoThinking significantly improves perception performance. The code, checkpoints, and dataset will be released after being accepted.
Authors: Hongyu Li, Kuan Liu, Yuan Chen, Juntao Hu, Huimin Lu, Guanjie Chen, Xue Liu, Guangming Lu, Hong Huang
Abstract: Recent advances in generative AI have demonstrated remarkable ability to produce high-quality content. However, these models often exhibit "Paradox of Simplicity": while they can render intricate landscapes, they often fail at simple, deterministic tasks. To address this, we formalize Obedience as the ability to align with instructions and establish a hierarchical grading system ranging from basic semantic alignment to pixel-level systemic precision, which provides a unified paradigm for incorporating and categorizing existing literature. Then, we conduct case studies to identify common obedience gaps, revealing how generative priors often override logical constraints. To evaluate high-level obedience, we present VIOLIN (VIsual Obedience Level-4 EvaluatIoN), the first benchmark focused on pure color generation across six variants. Extensive experiments on SOTA models reveal fundamental obedience limitations and further exploratory insights. By establishing this framework, we aim to draw more attention on AI Obedience and encourage deeper exploration to bridge this gap.
Authors: Irfan Atabas, Hatice Karatas
Abstract: In this study, image processing and deep learning methodologies were employed to automatically classify local olive species cultivated in Turkiye. A stereo camera was utilized to capture images of five distinct olive species, which were then preprocessed to ensure their suitability for analysis. Convolutional Neural Network (CNN) architectures, specifically MobileNetV2 and EfficientNetB0, were employed for image classification. These models were optimized through a transfer learning approach. The training and testing results indicated that the EfficientNetB0 model exhibited the optimal performance, with an accuracy of 94.5%. The findings demonstrate that deep learning-based systems offer an effective solution for classifying olive species with high accuracy. The developed method has significant potential for application in areas such as automatic identification and quality control of agricultural products.
Authors: Pr\'axedes Mart\'inez-Moreno, Andrea Valsecchi, Pablo Mesejo, Pilar Navarro-Ram\'irez, Valentino Lugli, Sergio Damas
Abstract: Craniofacial Superimposition is a forensic technique for identifying skeletal remains by comparing a post-mortem skull with ante-mortem facial photographs. A critical step in this process is Skull-Face Overlay (SFO). This stage involves aligning a 3D skull model with a 2D facial image, typically guided by cranial and facial landmarks' correspondence. However, its accuracy is undermined by individual variability in soft-tissue thickness, introducing significant uncertainty into the overlay. This paper introduces Lilium, an automated evolutionary method to enhance the accuracy and robustness of SFO. Lilium explicitly models soft-tissue variability using a 3D cone-based representation whose parameters are optimized via a Differential Evolution algorithm. The method enforces anatomical, morphological, and photographic plausibility through a combination of constraints: landmark matching, camera parameter consistency, head pose alignment, skull containment within facial boundaries, and region parallelism. This emulation of the usual forensic practitioners' approach leads Lilium to outperform the state-of-the-art method in terms of both accuracy and robustness.
Authors: Yuxiang Shen, Hailong Huang, Zhenkun Gao, Xueheng Li, Chengjun Xie, Xuanhua He, Jie Zhang
Abstract: Multimodal Large Language Models (MLLMs) are shifting towards "Thinking with Images" by actively exploring image details. While effective, large-scale training is computationally expensive, which has spurred growing interest in lightweight, training-free solutions. However, existing training-free methods suffer from two flaws: perceptual redundancy from indiscriminate cropping, which adds overhead and noise; and a drift between semantic intent and spatial attention, which prevents accurate localization of user-focused regions. To address these challenges, we propose AdaFocus, a novel training-free framework designed for adaptive visual reasoning. AdaFocus follows a two-stage pipeline: a confidence-based module decides when to crop, and a semantic-guided localization module determines where to crop. This enables adaptive visual reasoning without additional training. Experimentally, AdaFocus delivers substantial performance gains while achieving approximately 4.0\times speedup inference speedup than the SOTA method ZoomEyes, representing a significant advance in both accuracy and efficiency.
Authors: Simo Ryu, Chunghwan Han
Abstract: We describe our experience training Summer-22B, a video foundation model developed from scratch. This report documents the engineering challenges, design decisions, and lessons learned while scaling from raw footage collection to a functional model trained on approximately 50 million clips. We outline our approach combining metadata-driven dataset curation, multi-stage filtering, $\mu$P parameterization, and hypersphere-constrained optimization. We developed the Lavender Data system for dataset management and adopted inference-aware architectural choices. We share observations on what worked in our setting: dataset engineering consumed the majority of effort, architectural variants showed smaller differences than we expected, and $\mu$P hyperparameter transfer appeared effective even under geometric constraints. We hope this account proves useful to others undertaking similar projects.
Authors: Giorgio Roffo
Abstract: The quadratic cost of softmax attention limits Transformer scalability in high-resolution vision. We introduce Infinite Self-Attention (InfSA), a spectral reformulation that treats each attention layer as a diffusion step on a content-adaptive token graph, accumulating multi-hop interactions through a discounted Neumann series over attention matrices. This links self-attention to classical graph centrality (Katz, PageRank, eigenvector centrality) for interpretable token weighting. We also show the Neumann kernel equals the fundamental matrix of an absorbing Markov chain, so a token's centrality is its expected number of random-walk visits before absorption. We then propose Linear-InfSA, a linear-time variant that approximates the principal eigenvector of the implicit attention operator without forming the full attention matrix. It keeps an auxiliary state of fixed size proportional to per-head dimension dh (independent of sequence length N), is drop-in compatible with Vision Transformers, and supports stable training at 4096 by 4096 and inference at 9216 by 9216 (about 332k tokens). In a 4-layer ViT (53.5M parameters, 59 GFLOPs at 224 by 224), Linear-InfSA reaches 84.7% top-1 on ImageNet-1K, a +3.2 point architectural gain over an equal-depth softmax ViT trained with the same recipe. On ImageNet-V2, InfViT variants outperform all compared baselines (up to 79.8% vs 76.8%), indicating robustness under distribution shift. On an A100 40GB GPU, Linear-InfViT runs at 231 images/s and 0.87 J/image (13x better throughput and energy than equal-depth ViT) and is the only tested model to complete 9216 by 9216 inference without out-of-memory. The linear approximation closely matches the dominant eigenvector of the quadratic operator (cosine 0.985).
Authors: Abhinav Munagala
Abstract: Bird image segmentation remains a challenging task in computer vision due to extreme pose diversity, complex plumage patterns, and variable lighting conditions. This paper presents a dual-pipeline framework for binary bird image segmentation leveraging 2025 foundation models. We introduce two operating modes built upon Segment Anything Model 2.1 (SAM 2.1) as a shared frozen backbone: (1) a zero-shot pipeline using Grounding DINO 1.5 to detect birds via the text prompt "bird" before prompting SAM 2.1 with bounding boxes requiring no labelled bird data; and (2) a supervised pipeline that fine-tunes YOLOv11 on the CUB-200-2011 dataset for high-precision detection, again prompting SAM 2.1 for pixel-level masks. The segmentation model is never retrained for new species or domains. On CUB-200-2011 (11,788 images, 200 species), the supervised pipeline achieves IoU 0.912, Dice 0.954, and F1 0.953 outperforming all prior baselines including SegFormer-B2 (IoU 0.842) by +7.0 percentage points. The zero-shot pipeline achieves IoU 0.831 using only a text prompt, the first such result reported on this benchmark. We demonstrate that prompt-based foundation model pipelines outperform task specific end-to-end trained segmentation networks, while requiring only lightweight detector fine-tuning (~1 hour) for domain adaptation. Complete PyTorch implementation, dataset preparation scripts, and trained weights are publicly available.
Authors: Bowen Zhou, Zhou Xu, Wanli Li, Jingyu Xiao, Haoqian Wang
Abstract: Large Vision-Language Models (VLMs) have emerged as powerful engines for autonomous GUI agents, yet their deployment is severely constrained by the substantial memory footprint and latency of the Key-Value (KV) cache during long-horizon interactions. While existing cache compression methods have proven effective for LLMs, we empirically demonstrate that they suffer from suboptimal performance in GUI scenarios due to a fundamental misalignment: unlike general visual tasks where attention sparsity varies across layers, GUI attention patterns exhibit uniform high-sparsity across all transformer layers. Motivated by this insight, we propose ST-Lite, a training-free KV cache compression framework tailored for efficient GUI agents that explicitly addresses the dynamic spatio-trajectory dependencies within GUI data streams. ST-Lite introduces a novel dual-branch scoring policy incorporating Component-centric Spatial Saliency (CSS) and Trajectory-aware Semantic Gating (TSG). Specifically, CSS preserves the structural integrity of interactive UI elements by evaluating local neighborhood saliency, while TSG mitigates historical redundancy by dynamically filtering visually repetitive KV pairs within the interaction trajectory. Extensive evaluations demonstrate that with only a 10-20% cache budget, ST-Lite achieves a 2.45x decoding acceleration while maintaining comparable or even superior performance compared to full-cache baselines, offering a scalable solution for resource-constrained GUI agents.
Authors: Yang Yang, Xinze Zou, Zehua Ma, Han Fang, Weiming Zhang
Abstract: The rise of text-to-video generation models has raised growing concerns over content authenticity, copyright protection, and malicious misuse. Watermarking serves as an effective mechanism for regulating such AI-generated content, where high fidelity and strong robustness are particularly critical. Recent generative image watermarking methods provide a promising foundation by leveraging watermark information and pseudo-random keys to control the initial sampling noise, enabling lossless embedding. However, directly extending these techniques to videos introduces two key limitations: Existing designs implicitly rely on strict alignment between video frames and frame-dependent pseudo-random binary sequences used for watermark encryption. Once this alignment is disrupted, subsequent watermark extraction becomes unreliable; and Video-specific distortions, such as inter-frame compression, significantly degrade watermark reliability. To address these issues, we propose SKeDA, a generative watermarking framework tailored for text-to-video diffusion models. SKeDA consists of two components: (1) Shuffle-Key-based Distribution-preserving Sampling (SKe) employs a single base pseudo-random binary sequence for watermark encryption and derives frame-level encryption sequences through permutation. This design transforms watermark extraction from synchronization-sensitive sequence decoding into permutation-tolerant set-level aggregation, substantially improving robustness against frame reordering and loss; and (2) Differential Attention (DA), which computes inter-frame differences and dynamically adjusts attention weights during extraction, enhancing robustness against temporal distortions. Extensive experiments demonstrate that SKeDA preserves high video generation quality and watermark robustness.
Authors: Moumita Sen Sarma, Samatha Ereshi Akkamahadevi, Pascal Hitzler
Abstract: Deep Neural Networks (DNNs) have advanced applications in domains such as healthcare, autonomous systems, and scene understanding, yet the internal semantics of their hidden neurons remain poorly understood. Prior work introduced a Concept Induction-based framework for hidden neuron analysis and demonstrated its effectiveness on the ADE20K dataset. In this case study, we investigate whether the approach generalizes by applying it to the SUN2012 dataset, a large-scale scene recognition benchmark. Using the same workflow, we assign interpretable semantic labels to neurons and validate them through web-sourced images and statistical testing. Our findings confirm that the method transfers to SUN2012, showing its broader applicability.
Authors: Jindong Jiang, Amala Sanjay Deshmukh, Kateryna Chumachenko, Karan Sapra, Zhiding Yu, Guilin Liu, Andrew Tao, Pavlo Molchanov, Jan Kautz, Wonmin Byeon
Abstract: Token reduction is an effective way to accelerate long-video vision-language models (VLMs), but most existing methods are designed for dense Transformers and do not directly account for hybrid architectures that interleave attention with linear-time state-space blocks (e.g., Mamba). We study query-conditioned token reduction for hybrid video VLMs and analyze reduction behavior through two properties: layerwise sparsity (how many tokens capture query-relevant information) and importance stability (whether token-importance rankings persist across depth). Although token importance is sparse within each layer, the set of important tokens changes across layers, so aggressive early pruning is unreliable. Motivated by this, we propose a low-to-high progressive reduction schedule and a unified language-aware scoring mechanism for both attention and Mamba blocks (using an implicit-attention proxy for Mamba), enabling all-layer token reduction in hybrids. Under an aggressive compression setting (retaining 25% of visual tokens), our approach delivers substantial prefilling speedups (3.8--4.2x) with near-baseline accuracy at test time, and light finetuning under reduction further improves performance on long-context video benchmarks.
Authors: Antik Aich Roy, Ujjwal Bhattacharya
Abstract: One of the common issues in clinical decision-making is the presence of uncertainty, which often arises due to ambiguity in radiology reports, which often reflect genuine diagnostic uncertainty or limitations of automated label extraction in various complex cases. Especially the case of multilabel datasets such as CheXpert, MIMIC-CXR, etc., which contain labels such as positive, negative, and uncertain. In clinical decision-making, the uncertain label plays a tricky role as the model should not be forced to provide a confident prediction in the absence of sufficient evidence. The ability of the model to say it does not understand whenever it is not confident is crucial, especially in the cases of clinical decision-making involving high risks. Here, we propose AdURA-Net, a geometry-driven adaptive uncertainty-aware framework for reliable thoracic disease classification. The key highlights of the proposed model are: a) Adaptive dilated convolution and multiscale deformable alignment coupled with the backbone Densenet architecture capturing the anatomical complexities of the medical images, and b) Dual Head Loss, which combines masked binary cross entropy with logit and a Dirichlet evidential learning objective.
Authors: Daniel Nobrega Medeiros
Abstract: Existing visual reasoning benchmarks predominantly rely on natural language prompts, evaluate narrow reasoning modalities, or depend on subjective scoring procedures such as LLM-as-judge. We introduce the TACIT Benchmark, a programmatic visual reasoning benchmark comprising 10 tasks across 6 reasoning domains: spatial navigation, abstract pattern completion, causal simulation, logical constraint satisfaction, graph theory, and topology. The benchmark provides dual-track evaluation: a generative track in which models must produce solution images verified through deterministic computer-vision pipelines, and a discriminative track offering five-way multiple choice with structurally plausible near-miss distractors. Each distractor violates exactly one structural constraint, requiring models to reason about fine-grained visual differences rather than exploit superficial cues. Version 0.1.0 distributes 6,000 puzzles (108,000 PNG images across three resolutions) with fully deterministic seeded generation and reproducible verification. The dataset, generation code, and evaluation harness are released under the Apache 2.0 license on HuggingFace (DOI: 10.57967/hf/7904).
Authors: Soumya Suvra Ghosal, Youngeun Kim, Zhuowei Li, Ritwick Chaudhry, Linghan Xu, Hongjing Zhang, Jakub Zablocki, Yifan Xing, Qin Zhang
Abstract: Advances in large reasoning models have shown strong performance on complex reasoning tasks by scaling test-time compute through extended reasoning. However, recent studies observe that in vision-dependent tasks, extended textual reasoning at inference time can degrade performance as models progressively lose attention to visual tokens and increasingly rely on textual priors alone. To address this, prior works use reinforcement learning (RL)-based fine-tuning to route visual tokens or employ refocusing mechanisms during reasoning. While effective, these methods are computationally expensive, requiring large-scale data generation and policy optimization. To leverage the benefits of test-time compute without additional RL fine-tuning, we propose VisRef, a visually grounded test-time scaling framework. Our key idea is to actively guide the reasoning process by re-injecting a coreset of visual tokens that are semantically relevant to the reasoning context while remaining diverse and globally representative of the image, enabling more grounded multi-modal reasoning. Experiments on three visual reasoning benchmarks with state-of-the-art multi-modal large reasoning models demonstrate that, under fixed test-time compute budgets, VisRef consistently outperforms existing test-time scaling approaches by up to 6.4%.
Authors: Brianna D'Urso, Tahmid Hasan Sakib, Syed Rafay Hasan, Terry N. Guo
Abstract: This paper studies how well Naturalistic Adversarial Patches (NAPs) transfer to a physical traffic sign setting when the detector is trained on a customized dataset for an autonomous vehicle (AV) environment. We construct a composite dataset, CompGTSRB (which is customized dataset for AV environment), by pasting traffic sign instances from the German Traffic Sign Recognition Benchmark (GTSRB) onto undistorted backgrounds captured from the target platform. CompGTSRB is used to train a YOLOv5 model and generate patches using a Generative Adversarial Network (GAN) with latent space optimization, following existing NAP methods. We carried out a series of experiments on our Quanser QCar testbed utilizing the front CSI camera provided in QCar. Across configurations, NAPs reduce the detector's STOP class confidence. Different configurations include distance, patch sizes, and patch placement. These results along with a detailed step-by-step methodology indicate the utility of CompGTSRB dataset and the proposed systematic physical protocols for credible patch evaluation. The research further motivate researching the defenses that address localized patch corruption in embedded perception pipelines.
Authors: Giuseppe Sergioli, Carlo Cuccu, Giovanni Pasini, Alessandro Stefano, Giorgio Russo, Andr\'es Camilo Granda Arango, Roberto Giuntini
Abstract: We investigate a quantum-inspired approach to supervised multi-class classification based on the \emph{Pretty Good Measurement} (PGM), viewed as an operator-valued decision rule derived from quantum state discrimination. The method associates each class with an encoded mixed state and performs classification through a single POVM construction, thus providing a genuinely multi-class strategy without reduction to pairwise or one-vs-rest schemes. In this perspective, classification is reformulated as the discrimination of a finite ensemble of class-dependent density operators, with performance governed by the geometry induced by the encoding map and by the overlap structure among classes. To assess the practical scope of this framework, we apply the PGM-based classifier to two biomedical radiomics case studies: histopathological subtyping of non-small-cell lung carcinoma (NSCLC) and prostate cancer (PCa) risk stratification. The evaluation is conducted under protocols aligned with previously reported radiomics studies, enabling direct comparison with established classical baselines. The results show that the PGM-based classifier is consistently competitive and, in several settings, improves upon standard methods. In particular, the method performs especially well in the NSCLC binary and three-class tasks, while remaining competitive in the four-class case, where increased class overlap yields a more demanding discrimination geometry. In the PCa study, the PGM classifier remains close to the strongest ensemble baseline and exhibits clinically relevant sensitivity--specificity trade-offs across feature-selection scenarios.
Authors: He Li, Wenyue He, Weihang Kong, Xingchen Zhang
Abstract: Multimodal adversarial attacks for dense prediction remain largely underexplored. In particular, visual-infrared (VI) perception systems introduce unique challenges due to heterogeneous spectral characteristics and modality-specific intensity distributions. Existing adversarial patch methods are primarily designed for single-modal inputs and fail to account for crossspectral inconsistencies, leading to reduced attack effectiveness and poor stealthiness when applied to VI dense prediction models. To address these challenges, we propose a joint position-color optimization framework (AP-PCO) for generating adversarial patches in visual-infrared settings. The proposed method optimizes patch placement and color composition simultaneously using a fitness function derived from model outputs, enabling a single patch to perturb both visible and infrared modalities. To further bridge spectral discrepancies, we introduce a crossmodal color adaptation strategy that constrains patch appearance according to infrared grayscale characteristics while maintaining strong perturbations in the visible domain, thereby reducing cross-spectral saliency. The optimization procedure operates without requiring internal model information, supporting flexible black-box attacks. Extensive experiments on visual-infrared dense prediction tasks demonstrate that the proposed AP-PCO achieves consistently strong attack performance across multiple architectures, providing a practical benchmark for robustness evaluation in VI perception systems.
Authors: Unay Dorken Gallastegi, Wentao Shangguan, Vaibhav Choudhary, Akshay Agarwal, Hoover Rueda-Chac\'on, Martin J. Stevens, Vivek K Goyal
Abstract: Passive long-wave infrared (LWIR) absorption-based ranging relies on atmospheric absorption to estimate distances to objects from their emitted thermal radiation. First demonstrated decades ago for objects much hotter than the air and recently extended to scenes with low temperature variations, this ranging has depended on reflected radiance being negligible. Downwelling radiance is especially problematic, sometimes causing large inaccuracies. In two new ranging methods, we use characteristic features from ozone absorption to estimate the contribution of reflected downwelling radiance. The quadspectral method gives a simple closed-form range estimate from four narrowband measurements, two at a water vapor absorption line and two at an ozone absorption line. The hyperspectral method uses a broader spectral range to improve accuracy while also providing estimates of temperature, emissivity profiles, and contributions of downwelling from a collection of zenith angles. Experimental results demonstrate improved ranging accuracy, in one case reducing error from over 100 m when reflected light is not modeled to 6.8 m with the quadspectral method and 1.2 m with the hyperspectral method.
Authors: Boyu Chen, Weiye Bao, Junjie Liu, Michael Shen, Bo Peng, Paul Taylor, Zhu Li, Mengyue Yang
Abstract: Learning multimodal representations from medical images and other data sources can provide richer information for decision-making. While various multimodal models have been developed for this, they overlook learning features that are both necessary (must be present for the outcome to occur) and sufficient (enough to determine the outcome). We argue learning such features is crucial as they can improve model performance by capturing essential predictive information, and enhance model robustness to missing modalities as each modality can provide adequate predictive signals. Such features can be learned by leveraging the Probability of Necessity and Sufficiency (PNS) as a learning objective, an approach that has proven effective in unimodal settings. However, extending PNS to multimodal scenarios remains underexplored and is non-trivial as key conditions of PNS estimation are violated. We address this by decomposing multimodal representations into modality-invariant and modality-specific components, then deriving tractable PNS objectives for each. Experiments on synthetic and real-world medical datasets demonstrate our method's effectiveness. Code will be available on GitHub.
Authors: Arya Fayyazi, Haleh Akrami
Abstract: We present Proof-of-Perception (PoP), a tool-using framework that casts multimodal reasoning as an executable graph with explicit reliability guarantees. Each perception or logic node outputs a conformal set, yielding calibrated, stepwise uncertainty; a lightweight controller uses these certificates to allocate compute under a budget, expanding with extra tool calls only when needed and stopping early otherwise. This grounds answers in verifiable evidence, reduces error compounding and hallucinations, and enables principled accuracy-compute trade-offs. Across document, chart, and multi-image QA benchmarks, PoP improves performance and reliability over strong chain-of-thought, ReAct-style, and program-of-thought baselines while using computation more efficiently.
Authors: Xuanshuo Fu, Lei Kang, Javier Vazquez-Corral
Abstract: Low-light images often suffer from low contrast, noise, and color distortion, degrading visual quality and impairing downstream vision tasks. We propose a novel conditional diffusion framework for low-light image enhancement that incorporates a Structured Control Embedding Module (SCEM). SCEM decomposes a low-light image into four informative components including illumination, illumination-invariant features, shadow priors, and color-invariant cues. These components serve as control signals that condition a U-Net-based diffusion model trained with a simplified noise-prediction loss. Thus, the proposed SCEM equipped Diffusion method enforces structured enhancement guided by physical priors. In experiments, our model is trained only on the LOLv1 dataset and evaluated without fine-tuning on LOLv2-real, LSRW, DICM, MEF, and LIME. The method achieves state-of-the-art performance in quantitative and perceptual metrics, demonstrating strong generalization across benchmarks. https://casted.github.io/scem/.
Authors: Galen Pogoncheff, Alvin Wang, Jacob Granley, Michael Beyeler
Abstract: Cortical visual prostheses aim to restore sight by electrically stimulating neurons in early visual cortex (V1). With the emergence of high-density and flexible neural interfaces, electrode placement within three-dimensional cortex has become a critical surgical planning problem. Existing strategies emphasize visual field coverage and anatomical heuristics but do not directly optimize predicted perceptual outcomes under safety constraints. We present a percept-aware framework for surgical planning of cortical visual prostheses that formulates electrode placement as a constrained optimization problem in anatomical space. Electrode coordinates are treated as learnable parameters and optimized end-to-end using a differentiable forward model of prosthetic vision. The objective minimizes task-level perceptual error while incorporating vascular avoidance and gray matter feasibility constraints. Evaluated on simulated reading and natural image tasks using realistic folded cortical geometry (FreeSurfer fsaverage), percept-aware optimization consistently improves reconstruction fidelity relative to coverage-based placement strategies. Importantly, vascular safety constraints eliminate margin violations while preserving perceptual performance. The framework further enables co-optimization of multi-electrode thread configurations under fixed insertion budgets. These results demonstrate how differentiable percept models can inform anatomically grounded, safety-aware computer-assisted planning for cortical neural interfaces and provide a foundation for optimizing next-generation visual prostheses.
Authors: Austin Yunker, Peter Kenesei, Hemant Sharma, Jun-Sang Park, Antonino Miceli, Rajkumar Kettimuthu
Abstract: X-ray computed tomography (CT) is a widely used imaging technique that provides detailed examinations into the internal structure of an object with synchrotron CT (SR-CT) enabling improved data quality by using higher energy, monochromatic X-rays. While SR-CT allows for improved resolution, time-resolved experimentation, and reduced imaging artifacts, it also produces significantly larger datasets than conventional CT. Accurate and efficient evaluation of these datasets is a critical component of these workflows; yet is often done manually representing a major bottleneck in the analysis phase. While deep learning has emerged as a powerful tool capable of providing a wide range of purely data-driven solutions, it requires a substantial amount of labeled data for training and manual annotation of SR-CT datasets is impractical in practice. In this paper, we introduce a novel framework that enables automatic segmentation of large, high-resolution SR-CT datasets by eliminating the need to hand label images for deep learning training. First, we generate pseudo labels by clustering on the voxel values identifying regions in the volume with similar attenuation coefficients producing an initial semantic map. Afterwards, we train a segmentation model on the pseudo labels before utilizing the Unbiased Teacher approach to self-correct them ensuring accurate final segmentations. We find our approach improves pixel-wise accuracy and mIoU by 13.31% and 15.94%, respectively, over the baseline pseudo labels when using a magnesium crystal SR-CT sample. Additionally, we extensively evaluate the different components of our workflow including segmentation model, loss function, pseudo labeling strategy, and input type. Finally, we evaluate our approach on to two additional samples highlighting our frameworks ability to produce segmentations that are considerably better than the original pseudo labels.
Authors: Yujia Wu, Shuoqi Chen, Shiru Wang, Yucheng Tang, Petr Bruza, Geoffrey P. Luke
Abstract: Accurate Speed-of-Sound (SoS) reconstruction from acoustic waveforms is a cornerstone of ultrasound computed tomography (USCT), enabling quantitative velocity mapping that reveals subtle anatomical details and pathological variations often invisible in conventional imaging. However, practical utility is hindered by the limitations of existing algorithms; traditional Full Waveform Inversion (FWI) is computationally intensive, while current deep learning approaches tend to produce oversmoothed results lacking fine details. We propose DiffSOS, a conditional diffusion model that directly maps acoustic waveforms to SoS maps. Our framework employs a specialized acoustic ControlNet to strictly ground the denoising process in physical wave measurements. To ensure structural consistency, we optimize a hybrid loss function that integrates noise prediction, spatial reconstruction, and noise frequency content. To accelerate inference, we employ stochastic Denoising Diffusion Implicit Model (DDIM) sampling, achieving near real-time reconstruction with only 10 steps. Crucially, we exploit the stochastic generative nature of our framework to estimate pixel-wise uncertainty, providing a measure of reliability that is often absent in deterministic approaches. Evaluated on the OpenPros USCT benchmark, DiffSOS significantly outperforms state-of-the-art networks, achieving an average Multi-scale Structural Similarity of 0.957. Our approach provides high-fidelity SoS maps with a principled measure of confidence, facilitating safer and faster clinical interpretation.
Authors: Yi Zhang, Youya Xia, Yong Wang, Meng Song, Xin Wu, Wenjun Wan, Bingbing Liu, AiXue Ye, Hongbo Zhang, Feng Wen
Abstract: While Multimodal Large Language Models (MLLMs) excel in semantic tasks, they frequently lack the "spatial sense" essential for sophisticated geometric reasoning. Current models typically suffer from exorbitant modality-alignment costs and deficiency in fine-grained structural modeling precision.We introduce SSR, a framework designed for Structured Scene Reasoning that seamlessly integrates 2D and 3D representations via a lightweight alignment mechanism. To minimize training overhead, our framework anchors 3D geometric features to the large language model's pre-aligned 2D visual semantics through cross-modal addition and token interleaving, effectively obviating the necessity for large-scale alignment pre-training. To underpin complex spatial reasoning, we propose a novel scene graph generation pipeline that represents global layouts as a chain of independent local triplets defined by relative coordinates. This is complemented by an incremental generation algorithm, enabling the model to construct "language-model-friendly" structural scaffolds for complex environments. Furthermore, we extend these capabilities to global-scale 3D global grounding task, achieving absolute metric precision across heterogeneous data sources. At a 7B parameter scale, SSR achieves state-of-the-art performance on multiple spatial intelligence benchmarks, notably scoring 73.9 on VSI-Bench. Our approach significantly outperforms much larger models, demonstrating that efficient feature alignment and structured scene reasoning are the cornerstones of authentic spatial intelligence.
Authors: Yuanhao Su, Shaofeng Zhang, Xiaosong Jia, Qi Fan
Abstract: The development of 3D Vision-Language Models (VLMs), crucial for applications in robotics, autonomous driving, and augmented reality, is severely constrained by the scarcity of paired 3D-text data. Existing methods rely solely on next-token prediction loss, using only language tokens for supervision. This results in inefficient utilization of limited 3D data and leads to a significant degradation and loss of valuable geometric information in intermediate representations. To address these limitations, we propose {\mname}, a novel feature-level alignment regularization method. {\mname} explicitly supervises intermediate point cloud tokens to preserve fine-grained 3D geometric-semantic information throughout the language modeling process. Specifically, we constrain the intermediate point cloud tokens within the LLM to align with visual input tokens via a consistency loss. By training only a lightweight alignment projector and LoRA adapters, {\mname} achieves explicit feature-level supervision with minimal computational overhead, effectively preventing geometric degradation. Extensive experiments on ModelNet40 and Objaverse datasets demonstrate that our method achieves \textbf{2.08} pp improvement on average for classification tasks, with a substantial \textbf{7.50} pp gain on the challenging open-vocabulary Objaverse classification task and \textbf{4.88} pp improvement on 3D object captioning evaluated by Qwen2-72B-Instruct, validating the effectiveness of {\mname}. Code is publicly available at \href{https://github.com/yharoldsu0627/PointAlign}{https://github.com/yharoldsu0627/PointAlign}.
URLs: https://github.com/yharoldsu0627/PointAlign, https://github.com/yharoldsu0627/PointAlign
Authors: Changpu Li, Shuang Wu, Songlin Tang, Guangming Lu, Jun Yu, Wenjie Pei
Abstract: Reconstructing transparent objects from a set of multi-view images is a challenging task due to the complicated nature and indeterminate behavior of light propagation. Typical methods are primarily tailored to specific scenarios, such as objects following a uniform topology, exhibiting ideal transparency and surface specular reflections, or with only surface materials, which substantially constrains their practical applicability in real-world settings. In this work, we propose a differentiable rendering framework for transparent objects, dubbed DiffTrans, which allows for efficient decomposition and reconstruction of the geometry and materials of transparent objects, thereby reconstructing transparent objects accurately in intricate scenes with diverse topology and complex texture. Specifically, we first utilize FlexiCubes with dilation and smoothness regularization as the iso-surface representation to reconstruct an initial geometry efficiently from the multi-view object silhouette. Meanwhile, we employ the environment light radiance field to recover the environment of the scene. Then we devise a recursive differentiable ray tracer to further optimize the geometry, index of refraction and absorption rate simultaneously in a unified and end-to-end manner, leading to high-quality reconstruction of transparent objects in intricate scenes. A prominent advantage of the designed ray tracer is that it can be implemented in CUDA, enabling a significantly reduced computational cost. Extensive experiments on multiple benchmarks demonstrate the superior reconstruction performance of our DiffTrans compared with other methods, especially in intricate scenes involving transparent objects with diverse topology and complex texture. The code is available at https://github.com/lcp29/DiffTrans.
Authors: Doyi Kim, Minseok Seo, Changick Kim
Abstract: Precipitation forecasting relies on heterogeneous data. Weather radar is accurate, but coverage is geographically limited and costly to maintain. Weather stations provide accurate but sparse point measurements, while satellites offer dense, high-resolution coverage without direct rainfall retrieval. To overcome these limitations, we propose Query-Conditioned Gaussian Splatting (QCGS), the first framework to fuse automatic weather station (AWS) observations with satellite imagery for generating precipitation fields. Unlike conventional 2D Gaussian splatting, which renders the entire image plane, QCGS selectively renders only queried precipitation regions, avoiding unnecessary computation in non-precipitating areas while preserving sharp precipitation structures. The framework combines a radar point proposal network that identifies rainfall-support locations with an implicit neural representation (INR) network that predicts Gaussian parameters for each point. QCGS enables efficient, resolution-flexible precipitation field generation in real time. Through extensive evaluation with benchmark precipitation products, QCGS demonstrates over 50\% improvement in RMSE compared to conventional gridded precipitation products, and consistently maintains high performance across multiple spatiotemporal scales.
Authors: Hyungi Min, Taeseung You, Hangyeul Lee, Yeongjae Cho, Sungzoon Cho
Abstract: Counterfactual medical image generation have emerged as a critical tool for enhancing AI-driven systems in medical domain by answering "what-if" questions. However, existing approaches face two fundamental limitations: First, they fail to prevent unintended modifications, resulting collateral changes in demographic attributes when only disease features should be affected. Second, they lack interpretability in their editing process, which significantly limits their utility in real-world medical applications. To address these limitations, we present InstructX2X, a novel interpretable local editing model for counterfactual medical image generation featuring Region-Specific Editing. This approach restricts modifications to specific regions, effectively preventing unintended changes while simultaneously providing a Guidance Map that offers inherently interpretable visual explanations of the editing process. Additionally, we introduce MIMIC-EDIT-INSTRUCTION, a dataset for counterfactual medical image generation derived from expert-verified medical VQA pairs. Through extensive experiments, InstructX2X achieve state-of-the-art performance across all major evaluation metrics. Our model successfully generates high-quality counterfactual chest X-ray images along with interpretable explanations.
Authors: Hulingxiao He, Zhi Tan, Yuxin Peng
Abstract: A high-performing, general-purpose visual understanding model should map visual inputs to a taxonomic tree of labels, identify novel categories beyond the training set for which few or no publicly available images exist. Large Multimodal Models (LMMs) have achieved remarkable progress in fine-grained visual recognition (FGVR) for known categories. However, they remain limited in hierarchical visual recognition (HVR) that aims at predicting consistent label paths from coarse to fine categories, especially for novel categories. To tackle these challenges, we propose Taxonomy-Aware Representation Alignment (TARA), a simple yet effective strategy to inject taxonomic knowledge into LMMs. TARA leverages representations from biology foundation models (BFMs) that encode rich biological relationships through hierarchical contrastive learning. By aligning the intermediate representations of visual features with those of BFMs, LMMs are encouraged to extract discriminative visual cues well structured in the taxonomy tree. Additionally, we align the representations of the first answer token with the ground-truth label, flexibly bridging the gap between contextualized visual features and categories of varying granularity according to user intent. Experiments demonstrate that TARA consistently enhances LMMs' hierarchical consistency and leaf node accuracy, enabling reliable recognition of both known and novel categories within complex biological taxonomies. Code is available at https://github.com/PKU-ICST-MIPL/TARA_CVPR2026.
Authors: Hui Wan, Libin Lan
Abstract: Executing multiple tasks simultaneously in medical image analysis, including segmentation, classification, detection, and regression, often introduces significant challenges regarding model generalizability and the optimization of shared feature representations. While Vision Foundation Models (VFMs) provide powerful general representations, full fine-tuning on limited medical data is prone to overfitting and incurs high computational costs. Moreover, existing parameter-efficient fine-tuning approaches typically adopt task-agnostic adaptation protocols, overlooking both task-specific mechanisms and the varying sensitivity of model layers during fine-tuning. In this work, we propose Task-Aware Prompting and Selective Layer Fine-Tuning (TAP-SLF), a unified framework for multi-task ultrasound image analysis. TAP-SLF incorporates task-aware soft prompts to encode task-specific priors into the input token sequence and applies LoRA to selected specific top layers of the encoder. This strategy updates only a small fraction of the VFM parameters while keeping the pre-trained backbone frozen. By combining task-aware prompts with selective high-layer fine-tuning, TAP-SLF enables efficient VFM adaptation to diverse medical tasks within a shared backbone. Results on the FMC_UIA 2026 Challenge test set, where TAP-SLF wins fifth place, combined with evaluations on the officially released training dataset using an 8:2 train-test split, demonstrate that task-aware prompting and selective layer tuning are effective strategies for efficient VFM adaptation.
Authors: April Fu
Abstract: Although Large Vision-Language Models (LVLMs) have made substantial progress, hallucination, where generated text is not grounded in the visual input, remains a challenge. As LVLMs become stronger, previously reported hallucination patterns, such as linguistic bias and overthinking phenomenon, become far less consistent, making the corresponding mitigation techniques substantially less effective. In this paper, we introduce an Internal self-Correction mechanism utilizing Layer Attention (ICLA) that operates directly on hidden states during generation. Each layer selectively retrieves information from all preceding layers through a diagonal cross-layer attention mechanism, enabling self-refinement without any external correction signals. With introducing and training only 0.2M and 0.1M additional parameters on LLaVA1.5-7B and Qwen2.5-VL-7B, \ours consistently improves visual grounding across multiple hallucination benchmarks, demonstrating its effectiveness for more advanced LVLMs.
Authors: Xueyang Li, Yunzhong Lou, Yu Song, Xiangdong Zhou
Abstract: Computer-Aided Design (CAD) generative modeling has a strong and long-term application in the industry. Recently, the parametric CAD sequence as the design logic of an object has been widely mined by sequence models. However, the industrial CAD models, especially in component objects, are fine-grained and complex, requiring a longer parametric CAD sequence to define. To address the problem, we introduce Mamba-CAD, a self-supervised generative modeling for complex CAD models in the industry, which can model on a longer parametric CAD sequence. Specifically, we first design an encoder-decoder framework based on a Mamba architecture and pair it with a CAD reconstruction task for pre-training to model the latent representation of CAD models; and then we utilize the learned representation to guide a generative adversarial network to produce the fake representation of CAD models, which would be finally recovered into parametric CAD sequences via the decoder of MambaCAD. To train Mamba-CAD, we further create a new dataset consisting of 77,078 CAD models with longer parametric CAD sequences. Comprehensive experiments are conducted to demonstrate the effectiveness of our model under various evaluation metrics, especially in the generation length of valid parametric CAD sequences. The code and dataset can be achieved from https://github.com/Sunny-Hack/Code-for-Mamba-CAD-AAAI-2025-.
URLs: https://github.com/Sunny-Hack/Code-for-Mamba-CAD-AAAI-2025-.
Authors: Zhuoran Zhao, Xianghao Kong, Linlin Yang, Zheng Wei, Pan Hui, Anyi Rao
Abstract: Recent studies on 3D hand reconstruction have demonstrated the effectiveness of synthetic training data to improve estimation performance. However, most methods rely on game engines to synthesize hand images, which often lack diversity in textures and environments, and fail to include crucial components like arms or interacting objects. Generative models are promising alternatives to generate diverse hand images, but still suffer from misalignment issues. In this paper, we present SesaHand, which enhances controllable hand image generation from both semantic and structural alignment perspectives for 3D hand reconstruction. Specifically, for semantic alignment, we propose a pipeline with Chain-of-Thought inference to extract human behavior semantics from image captions generated by the Vision-Language Model. This semantics suppresses human-irrelevant environmental details and ensures sufficient human-centric contexts for hand image generation. For structural alignment, we introduce hierarchical structural fusion to integrate structural information with different granularity for feature refinement to better align the hand and the overall human body in generated images. We further propose a hand structure attention enhancement method to efficiently enhance the model's attention on hand regions. Experiments demonstrate that our method not only outperforms prior work in generation performance but also improves 3D hand reconstruction with the generated hand images.
Authors: Bin Chen, Weiqi Li, Shijie Zhao, Xuanyu Zhang, Junlin Li, Li Zhang, Jian Zhang
Abstract: While many diffusion models have achieved impressive results in real-world video super-resolution (Real-VSR) by generating rich and realistic details, their reliance on multi-step sampling leads to slow inference. One-step networks like SeedVR2, DOVE, and DLoRAL alleviate this through condensing generation into one single step, yet they remain heavy, with billions of parameters and multi-second latency. Recent adversarial diffusion compression (ADC) offers a promising path via pruning and distilling these models into a compact AdcSR network, but directly applying it to Real-VSR fails to balance spatial details and temporal consistency due to its lack of temporal awareness and the limitations of standard adversarial learning. To address these challenges, we propose an improved ADC method for Real-VSR. Our approach distills a large diffusion Transformer (DiT) teacher DOVE equipped with 3D spatio-temporal attentions, into a pruned 2D Stable Diffusion (SD)-based AdcSR backbone, augmented with lightweight 1D temporal convolutions, achieving significantly higher efficiency. In addition, we introduce a dual-head adversarial distillation scheme, in which discriminators in both pixel and feature domains explicitly disentangle the discrimination of details and consistency into two heads, enabling both objectives to be effectively optimized without sacrificing one for the other. Experiments demonstrate that the resulting compressed AdcVSR model reduces complexity by 95% in parameters and achieves an 8$\times$ acceleration over its DiT teacher DOVE, while maintaining competitive video quality and efficiency.
Authors: Rajdeep Chatterjee, Sudip Chakrabarty, Trishaani Acharjee
Abstract: Accurate semantic segmentation of foot ulcers is essential for automated wound monitoring, yet boundary delineation remains challenging due to tissue heterogeneity and poor contrast with surrounding skin. To overcome the limitations of standard intensity-based networks, we present LSS-LTCNet:an ante-hoc explainable framework synergizing deterministic structural priors with continuous-time neural dynamics. Our architecture departs from traditional black-box models by employing a Local Self-Similarity (LSS) mechanism that extracts dense, illumination-invariant texture descriptors to explicitly disentangle necrotic tissue from background artifacts. To enforce topological precision, we introduce a Liquid Time-Constant (LTC) refinement module that treats boundary evolution as an ODEgoverned dynamic system, iteratively refining masks over continuous time-steps. Comprehensive evaluation on the MICCAI FUSeg dataset demonstrates that LSS-LTCNet achieves state-of-the-art boundary alignment, securing a peak Dice score of 86.96% and an exceptional 95th percentile Hausdorff Distance (HD95) of 8.91 pixels. Requiring merely 25.70M parameters, the model significantly outperforms heavier U-Net and transformer baselines in efficiency. By providing inherent visual audit trails alongside high-fidelity predictions, LSS-LTCNet offers a robust and transparent solution for computer-aided diagnosis in mobile healthcare (mHealth) settings.
Authors: Cong Wan, Zeyu Guo, Jiangyang Li, SongLin Dong, Yifan Bai, Lin Peng, Zhiheng Ma, Yihong Gong
Abstract: We present ReMoT, a unified training paradigm to systematically address the fundamental shortcomings of VLMs in spatio-temporal consistency -- a critical failure point in navigation, robotics, and autonomous driving. ReMoT integrates two core components: (1) A rule-based automatic framework that generates ReMoT-16K, a large-scale (16.5K triplets) motion-contrast dataset derived from video meta-annotations, surpassing costly manual or model-based generation. (2) Group Relative Policy Optimization, which we empirically validate yields optimal performance and data efficiency for learning this contrastive reasoning, far exceeding standard Supervised Fine-Tuning. We also construct the first benchmark for fine-grained motion contrast triplets to measure a VLM's discrimination of subtle motion attributes (e.g., opposing directions). The resulting model achieves state-of-the-art performance on our new benchmark and multiple standard VLM benchmarks, culminating in a remarkable 25.1% performance leap on spatio-temporal reasoning tasks.
Authors: Zhaolin Yu, Litao Yang, Ben Babicka, Ming Hu, Jing Hao, Anthony Huang, James Huang, Yueming Jin, Jiasong Wu, Zongyuan Ge
Abstract: Orthopantomograms (OPGs) are the standard panoramic radiograph in dentistry, used for full-arch screening across multiple diagnostic tasks. While Vision Language Models (VLMs) now allow multi-task OPG analysis through natural language, they underperform task-specific models on most individual tasks. Agentic systems that orchestrate specialized tools offer a path to both versatility and accuracy, this approach remains unexplored in the field of dental imaging. To address this gap, we propose OPGAgent, a multi-tool agentic system for auditable OPG interpretation. OPGAgent coordinates specialized perception modules with a consensus mechanism through three components: (1) a Hierarchical Evidence Gathering module that decomposes OPG analysis into global, quadrant, and tooth-level phases with dynamically invoking tools, (2) a Specialized Toolbox encapsulating spatial, detection, utility, and expert zoos, and (3) a Consensus Subagent that resolves conflicts through anatomical constraints. We further propose OPG-Bench, a structured-report protocol based on (Location, Field, Value) triples derived from real clinical reports, which enables a comprehensive review of findings and hallucinations, extending beyond the limitations of VQA indicators. On our OPG-Bench and the public MMOral-OPG benchmark, OPGAgent outperforms current dental VLMs and medical agent frameworks across both structured-report and VQA evaluation. Code will be released upon acceptance.
Authors: Boming Tan, Xiangdong Zhang, Ning Liao, Yuqing Zhang, Shaofeng Zhang, Xue Yang, Qi Fan, Yanyong Zhang
Abstract: Despite impressive progress in video generation, existing models remain limited to surface-level plausibility, lacking a coherent and unified understanding of the world. Prior approaches typically incorporate only a single form of world-related knowledge or rely on rigid alignment strategies to introduce additional knowledge. However, aligning the single world knowledge is insufficient to constitute a world model that requires jointly modeling multiple heterogeneous dimensions (e.g., physical commonsense, 3D and temporal consistency). To address this limitation, we introduce \textbf{DreamWorld}, a unified framework that integrates complementary world knowledge into video generators via a \textbf{Joint World Modeling Paradigm}, jointly predicting video pixels and features from foundation models to capture temporal dynamics, spatial geometry, and semantic consistency. However, naively optimizing these heterogeneous objectives can lead to visual instability and temporal flickering. To mitigate this issue, we propose \textit{Consistent Constraint Annealing (CCA)} to progressively regulate world-level constraints during training, and \textit{Multi-Source Inner-Guidance} to enforce learned world priors at inference. Extensive evaluations show that DreamWorld improves world consistency, outperforming Wan2.1 by 2.26 points on VBench. Code will be made publicly available at \href{https://github.com/ABU121111/DreamWorld}{\textcolor{mypink}{\textbf{Github}}}.
Authors: Pengju Sun, Banglei Guan, Jing Tao, Zhenbao Yu, Xuanyu Bai, Yang Shang, Qifeng Yu
Abstract: High dynamic range (HDR) imaging under extreme illumination remains challenging for conventional cameras due to overexposure. Event cameras provide microsecond temporal resolution and high dynamic range, while spatially varying exposure (SVE) sensors offer single-shot radiometric diversity.We present a hardware--algorithm co-designed HDR imaging system that tightly integrates an SVE micro-attenuation camera with an event sensor in an asymmetric dual-modality configuration. To handle non-coaxial geometry and heterogeneous optics, we develop a two-stage cross-modal alignment framework that combines feature-guided coarse homography estimation with a multi-scale refinement module based on spatial pooling and frequency-domain filtering. On top of aligned representations, we develop a cross-modal HDR reconstruction network with convolutional fusion, mutual-information regularization, and a learnable fusion loss that adaptively balances intensity cues and event-derived structural constraints. Comprehensive experiments on both synthetic benchmarks and real captures demonstrate that the proposed system consistently improves highlight recovery, edge fidelity, and robustness compared with frame-only or event-only HDR pipelines. The results indicate that jointly optimizing optical design, cross-modal alignment, and computational fusion provides an effective foundation for reliable HDR perception in highly dynamic and radiometrically challenging environments.
Authors: Pengcheng Shi, Minghui Zhang, Kehan Song, Jiaqi Liu, Yun Gu, Xinglin Zhang
Abstract: Automated radiology report generation is key for reducing radiologist workload and improving diagnostic consistency, yet generating accurate reports for 3D medical imaging remains challenging. Existing vision-language models face two limitations: they do not leverage segmentation-pretrained encoders, and they inject visual features only at the input layer of language models, losing multi-scale information. We propose U-VLM, which enables hierarchical vision-language modeling in both training and architecture: (1) progressive training from segmentation to classification to report generation, and (2) multi-layer visual injection that routes U-Net encoder features to corresponding language model layers. Each training stage can leverage different datasets without unified annotations. U-VLM achieves state-of-the-art performance on CT-RATE (F1: 0.414 vs 0.258, BLEU-mean: 0.349 vs 0.305) and AbdomenAtlas 3.0 (F1: 0.624 vs 0.518 for segmentation-based detection) using only a 0.1B decoder trained from scratch, demonstrating that well-designed vision encoder pretraining outweighs the benefits of 7B+ pre-trained language models. Ablation studies show that progressive pretraining significantly improves F1, while multi-layer injection improves BLEU-mean. Code is available at https://github.com/yinghemedical/U-VLM.
Authors: Feibo Jiang, Siwei Tu, Li Dong, Xiaolong Li, Kezhi Wang, Cunhua Pan, Zhu Han, Jiangzhou Wang
Abstract: Visual-Language Models (VLMs), with their strong capabilities in image and text understanding, offer a solid foundation for intelligent communications. However, their effectiveness is constrained by limited token granularity, overlong visual token sequences, and inadequate cross-modal alignment. To overcome these challenges, we propose TaiChi, a novel VLM framework designed for token communications. TaiChi adopts a dual-visual tokenizer architecture that processes both high- and low-resolution images to collaboratively capture pixel-level details and global conceptual features. A Bilateral Attention Network (BAN) is introduced to intelligently fuse multi-scale visual tokens, thereby enhancing visual understanding and producing compact visual tokens. In addition, a Kolmogorov Arnold Network (KAN)-based modality projector with learnable activation functions is employed to achieve precise nonlinear alignment from visual features to the text semantic space, thus minimizing information loss. Finally, TaiChi is integrated into a multimodal and multitask token communication system equipped with a joint VLM-channel coding scheme. Experimental results validate the superior performance of TaiChi, as well as the feasibility and effectiveness of the TaiChi-driven token communication system.
Authors: Liyao Jiang, Ruichen Chen, Chao Gao, Di Niu
Abstract: Recent text-to-image (T2I) diffusion models achieve remarkable realism, yet faithful prompt-image alignment remains challenging, particularly for complex prompts with multiple objects, relations, and fine-grained attributes. Existing training-free inference-time scaling methods rely on fixed iteration budgets that cannot adapt to prompt difficulty, while reflection-tuned models require carefully curated reflection datasets and extensive joint fine-tuning of diffusion and vision-language models, often overfitting to reflection paths data and lacking transferability across models. We introduce RAISE (Requirement-Adaptive Self-Improving Evolution), a training-free, requirement-driven evolutionary framework for adaptive T2I generation. RAISE formulates image generation as a requirement-driven adaptive scaling process, evolving a population of candidates at inference time through a diverse set of refinement actions-including prompt rewriting, noise resampling, and instructional editing. Each generation is verified against a structured checklist of requirements, enabling the system to dynamically identify unsatisfied items and allocate further computation only where needed. This achieves adaptive test-time scaling that aligns computational effort with semantic query complexity. On GenEval and DrawBench, RAISE attains state-of-the-art alignment (0.94 overall GenEval) while incurring fewer generated samples (reduced by 30-40%) and VLM calls (reduced by 80%) than prior scaling and reflection-tuned baselines, demonstrating efficient, generalizable, and model-agnostic multi-round self-improvement. Code is available at https://github.com/LiyaoJiang1998/RAISE.
Authors: Qihang Fan, Yuang Ai, Huaibo Huang, Ran He
Abstract: Since Transformers are introduced into vision architectures, their quadratic complexity has always been a significant issue that many research efforts aim to address. A representative approach involves grouping tokens, performing self-attention calculations within each group, or pooling the tokens within each group into a single token. To this end, various carefully designed grouping strategies have been proposed to enhance the performance of Vision Transformers. Here, we pose the following questions: \textbf{Are these carefully designed grouping methods truly necessary? Is there a simpler and more unified token grouping method that can replace these diverse methods?} Therefore, we propose the random grouping strategy, which involves a simple and fast random grouping strategy for vision tokens. We validate this approach on multiple baselines, and experiments show that random grouping almost outperforms all other grouping methods. When transferred to downstream tasks, such as object detection, random grouping demonstrates even more pronounced advantages. In response to this phenomenon, we conduct a detailed analysis of the advantages of random grouping from multiple perspectives and identify several crucial elements for the design of grouping strategies: positional information, head feature diversity, global receptive field, and fixed grouping pattern. We demonstrate that as long as these four conditions are met, vision tokens require only an extremely simple grouping strategy to efficiently and effectively handle various visual tasks. We also validate the effectiveness of our proposed random method across multiple modalities, including visual tasks, point cloud processing, and vision-language models. Code will be available at https://github.com/qhfan/random.
Authors: Riccardo de Lutio, Tobias Fischer, Yen-Yu Chang, Yuxuan Zhang, Jay Zhangjie Wu, Xuanchi Ren, Tianchang Shen, Katarina Tothova, Zan Gojcic, Haithem Turki
Abstract: Per-scene optimization methods such as 3D Gaussian Splatting provide state-of-the-art novel view synthesis quality but extrapolate poorly to under-observed areas. Methods that leverage generative priors to correct artifacts in these areas hold promise but currently suffer from two shortcomings. The first is scalability, as existing methods use image diffusion models or bidirectional video models that are limited in the number of views they can generate in a single pass (and thus require a costly iterative distillation process for consistency). The second is quality itself, as generators used in prior work tend to produce outputs that are inconsistent with existing scene content and fail entirely in completely unobserved regions. To solve these, we propose a two-stage pipeline that leverages two key insights. First, we train a powerful bidirectional generative model with a novel opacity mixing strategy that encourages consistency with existing observations while retaining the model's ability to extrapolate novel content in unseen areas. Second, we distill it into a causal auto-regressive model that generates hundreds of frames in a single pass. This model can directly produce novel views or serve as pseudo-supervision to improve the underlying 3D representation in a simple and highly efficient manner. We evaluate our method extensively and demonstrate that it can generate plausible reconstructions in scenarios where existing approaches fail completely. When measured on commonly benchmarked datasets, we outperform existing all existing baselines by a wide margin, exceeding prior state-of-the-art methods by 1-3 dB PSNR.
Authors: Yuchen Che, Jingtu Wu, Hao Zheng, Asako Kanezaki
Abstract: Estimating the 6DoF pose of a novel object with a single reference view is challenging due to occlusions, view-point changes, and outliers. A core difficulty lies in finding robust cross-view correspondences, as existing methods often rely on discrete one-to-one matching that is non-differentiable and tends to collapse onto sparse key-points. We propose Confidence-aware Optimal Geometric Correspondence (COG), an unsupervised framework that formulates correspondence estimation as a confidence-aware optimal transport problem. COG produces balanced soft correspondences by predicting point-wise confidences and injecting them as optimal transport marginals, suppressing non-overlapping regions. Semantic priors from vision foundation models further regularize the correspondences, leading to stable pose estimation. This design integrates confidence into the correspondence finding and pose estimation pipeline, enabling unsupervised learning. Experiments show unsupervised COG achieves comparable performance to supervised methods, and supervised COG outperforms them.
Authors: Dawei Yan, Haokui Zhang, Guangda Huzhang, Yang Li, Yibo Wang, Qing-Guo Chen, Zhao Xu, Weihua Luo, Ying Li, Wei Dong, Chunhua Shen
Abstract: Multimodal Large Language Models (MLLMs) based agents have demonstrated remarkable potential in autonomous web navigation. However, handling long-horizon tasks remains a critical bottleneck. Prevailing strategies often rely heavily on extensive data collection and model training, yet still struggle with high computational costs and insufficient reasoning capabilities when facing complex, long-horizon scenarios. To address this, we propose M$^2$, a training-free, memory-augmented framework designed to optimize context efficiency and decision-making robustness. Our approach incorporates a dual-tier memory mechanism that synergizes Dynamic Trajectory Summarization (Internal Memory) to compress verbose interaction history into concise state updates, and Insight Retrieval Augmentation (External Memory) to guide the agent with actionable guidelines retrieved from an offline insight bank. Extensive evaluations across WebVoyager and OnlineMind2Web demonstrate that M$^2$ consistently surpasses baselines, yielding up to a 19.6% success rate increase and 58.7% token reduction for Qwen3-VL-32B, while proprietary models like Claude achieve accuracy gains up to 12.5% alongside significantly lower computational overhead.
Authors: Keunho Byeon, Jinsol Song, Seong Min Hong, Yosep Chong, Jin Tae Kwak
Abstract: Whole-slide image analysis is essential for diagnostic tasks in pathology, yet existing deep learning methods primarily rely on flat classification, ignoring hierarchical relationships among class labels. In this study, we propose HiClass, a hierarchical classification framework for improved histopathology image analysis, that enhances both coarse-grained and fine-grained WSI classification. Built based upon a multiple instance learning approach, HiClass extends it by introducing bidirectional feature integration that facilitates information exchange between coarse-grained and fine-grained feature representations, effectively learning hierarchical features. Moreover, we introduce tailored loss functions, including hierarchical consistency loss, intra- and inter-class distance loss, and group-wise cross-entropy loss, to further optimize hierarchical learning. We assess the performance of HiClass on a gastric biopsy dataset with 4 coarse-grained and 14 fine-grained classes, achieving superior classification performance for both coarse-grained classification and fine-grained classification. These results demonstrate the effectiveness of HiClass in improving WSI classification by capturing coarse-grained and fine-grained histopathological characteristics.
Authors: Yingqi Fan, Junlong Tong, Anhao Zhao, Xiaoyu Shen
Abstract: Multimodal large language models (MLLMs) project visual tokens into the embedding space of language models, yet the internal structuring and processing of visual semantics remain poorly understood. In this work, we introduce a two-fold analytical framework featuring a novel probing tool, $\textbf{EmbedLens}$, to conduct a fine-grained analysis. We uncover a pronounced semantic sparsity at the input level: visual tokens consistently partition into sink, dead, and alive categories. Remarkably, only the alive tokens, comprising $\approx60\%$ of the total input, carry image-specific meaning. Furthermore, using a targeted patch-compression benchmark, we demonstrate that these alive tokens already encode rich, fine-grained cues (e.g., objects, colors, and OCR) prior to entering the LLM. Internal visual computations (such as visual attention and feed-forward networks) are redundant for most standard tasks. For the small subset of highly vision-centric tasks that actually benefit from internal processing, we reveal that alive tokens naturally align with intermediate LLM layers rather than the initial embedding space, indicating that shallow-layer processing is unnecessary and that direct mid-layer injection is both sufficient. Ultimately, our findings provide a unified mechanistic view of visual token processing, paving the way for more efficient and interpretable MLLM architectures through selective token pruning, minimized visual computation, and mid-layer injection. The code is released at: https://github.com/EIT-NLP/EmbedLens.
Authors: Ruoshuang Du, Xin Sun, Qiang Liu, Bowen Song, Zhongqi Chen, Weiqiang Wang, Liang Wang
Abstract: Visual Question Answering systems face reliability issues due to hallucinations, where models generate answers misaligned with visual input or factual knowledge. While Retrieval Augmented Generation frameworks mitigate this issue by incorporating external knowledge, static retrieval often introduces irrelevant or conflicting content, particularly in visual RAG settings where visually similar but semantically incorrect evidence may be retrieved. To address this, we propose Multimodal Adaptive RAG (MMA-RAG), which dynamically assesses the confidence in the internal knowledge of the model to decide whether to incorporate the retrieved external information into the generation process. Central to MMA-RAG is a decision classifier trained through a layer-wise analysis, which leverages joint internal visual and textual representations to guide the use of reverse image retrieval. Experiments demonstrated that the model achieves a significant improvement in response performance in three VQA datasets. Meanwhile, ablation studies highlighted the importance of internal representations in adaptive retrieval decisions. In general, the experimental results demonstrated that MMA-RAG effectively balances external knowledge utilization and inference robustness in diverse multimodal scenarios.
Authors: Wang Chen, Yuhui Zeng, Yongdong Luo, Tianyu Xie, Luojun Lin, Jiayi Ji, Yan Zhang, Xiawu Zheng
Abstract: Frame selection is crucial due to high frame redundancy and limited context windows when applying Large Vision-Language Models (LVLMs) to long videos. Current methods typically select frames with high relevance to a given query, resulting in a disjointed set of frames that disregard the narrative structure of video. In this paper, we introduce Wavelet-based Frame Selection by Detecting Semantic Boundary (WFS-SB), a training-free framework that presents a new perspective: effective video understanding hinges not only on high relevance but, more importantly, on capturing semantic shifts - pivotal moments of narrative change that are essential to comprehending the holistic storyline of video. However, direct detection of abrupt changes in the query-frame similarity signal is often unreliable due to high-frequency noise arising from model uncertainty and transient visual variations. To address this, we leverage the wavelet transform, which provides an ideal solution through its multi-resolution analysis in both time and frequency domains. By applying this transform, we decompose the noisy signal into multiple scales and extract a clean semantic change signal from the coarsest scale. We identify the local extrema of this signal as semantic boundaries, which segment the video into coherent clips. Building on this, WFS-SB comprises a two-stage strategy: first, adaptively allocating a frame budget to each clip based on a composite importance score; and second, within each clip, employing the Maximal Marginal Relevance approach to select a diverse yet relevant set of frames. Extensive experiments show that WFS-SB significantly boosts LVLM performance, e.g., improving accuracy by 5.5% on VideoMME, 9.5% on MLVU, and 6.2% on LongVideoBench, consistently outperforming state-of-the-art methods.
Authors: Xingyilang Yin, Chengzhengxu Li, Jiahao Chang, Chi-Man Pun, Xiaodong Cun
Abstract: Humans are born with vision-based 4D spatial-temporal intelligence, which enables us to perceive and reason about the evolution of 3D space over time from purely visual inputs. Despite its importance, this capability remains a significant bottleneck for current multimodal large language models (MLLMs). To tackle this challenge, we introduce MLLM-4D, a comprehensive framework designed to bridge the gaps in training data curation and model post-training for spatiotemporal understanding and reasoning. On the data front, we develop a cost-efficient data curation pipeline that repurposes existing stereo video datasets into high-quality 4D spatiotemporal instructional data. This results in the MLLM4D-2M and MLLM4D-R1-30k datasets for Supervised Fine-Tuning (SFT) and Reinforcement Fine-Tuning (RFT), alongside MLLM4D-Bench for comprehensive evaluation. Regarding model training, our post-training strategy establishes a foundational 4D understanding via SFT and further catalyzes 4D reasoning capabilities by employing Group Relative Policy Optimization (GRPO) with specialized Spatiotemporal Chain of Thought (ST-CoT) prompting and Spatiotemporal reward functions (ST-reward) without involving the modification of architecture. Extensive experiments demonstrate that MLLM-4D achieves state-of-the-art spatial-temporal understanding and reasoning capabilities from purely 2D RGB inputs. Project page: https://github.com/GVCLab/MLLM-4D.
Authors: Quan Kong, Yanru Xiao, Yuhao Shen, Cong Wang
Abstract: Learning efficient and expressive visual representation has long been the pursuit of computer vision research. While Vision Transformers (ViTs) gradually replace traditional Convolutional Neural Networks (CNNs) as more scalable vision learners, their applications are plagued by the quadratic complexity of the self-attention mechanism. To address the challenge, we introduce a new linear-time sequence modeling method Test-Time Training (TTT) into vision and propose Vision-TTT, which compresses the visual token sequence in a novel self-supervised learning manner. By incorporating bidirectional scan strategy and the Conv2d module, Vision-TTT effectively extends vanilla TTT to model 2D visual correlations with global receptive fields. Extensive experiments show that \texttt{Vittt-T/S/B} achieve 77.3%,81.2%,82.5% Top-1 accuracy on ImageNet classification and also greatly outperform their counterparts on downstream tasks. At 1280x1280 resolution, \texttt{Vittt-T} reduces FLOPs by 79.4% and runs 4.38x faster with 88.9% less memory than DeiT-T. These results demonstrate the expressiveness and efficiency of Vision-TTT as a strong candidate for the next-generation generic visual backbone.
Authors: Yuyang Chen, Linqian Zeng, Yijin ZHou, Hengjie Li, Jidong Zhai
Abstract: Diffusion models have achieved remarkable success in generative AI, yet their computational efficiency remains a significant challenge, particularly for Diffusion Transformers (DiTs) requiring intensive full-attention computation. While existing acceleration approaches focus on content-agnostic uniform optimization strategies, we observe that different regions in generated content exhibit heterogeneous convergence patterns during the denoising process. We present Jano, a training-free framework that leverages this insight for efficient region-aware generation. Jano introduces an early-stage complexity recognition algorithm that accurately identifies regional convergence requirements within initial denoising steps, coupled with an adaptive token scheduling runtime that optimizes computational resource allocation. Through comprehensive evaluation on state-of-the-art models, Jano achieves substantial acceleration (average 2.0 times speedup, up to 2.4 times) while preserving generation quality. Our work challenges conventional uniform processing assumptions and provides a practical solution for accelerating large-scale content generation. The source code of our implementation is available at https://github.com/chen-yy20/Jano.
Authors: Zhen Zhou, Jian Liu, Biwen Lei, Jing Xu, Haohan Weng, Yiling Zhu, Zhuo Chen, Junfeng Fan, Yunkai Ma, Dazhao Du, Song Guo, Fengshui Jing, Chunchao Guo
Abstract: Reinforcement learning (RL) has demonstrated remarkable success in text and image generation, yet its potential in 3D generation remains largely unexplored. Existing attempts typically rely on offline direct preference optimization (DPO) method, which suffers from low training efficiency and limited generalization. In this work, we aim to enhance both the training efficiency and generation quality of RL in 3D mesh generation. Specifically, (1) we design the first asynchronous online RL framework tailored for 3D mesh generation post-training efficiency improvement, which is 3.75$\times$ faster than synchronous RL. (2) We propose Advantage-guided Ranking Preference Optimization (ARPO), a novel RL algorithm that achieves a better trade-off between training efficiency and generalization than current RL algorithms designed for 3D mesh generation, such as DPO and group relative policy optimization (GRPO). (3) Based on asynchronous ARPO, we propose Mesh-Pro, which additionally introduces a novel diagonal-aware mixed triangular-quadrilateral tokenization for mesh representation and a ray-based reward for geometric integrity. Mesh-Pro achieves state-of-the-art performance on artistic and dense meshes.
Authors: Wenjie Wei, Xiaolong Zhou, Malu Zhang, Ammar Belatreche, Qian Sun, Yimeng Shan, Dehao Zhang, Zijian Zhou, Zeyu Ma, Yang Yang, Haizhou Li
Abstract: Spiking neural networks (SNNs) offer an energy-efficient alternative to traditional neural networks due to their event-driven computing paradigm. However, recent advancements in spiking transformers have focused on improving accuracy with large-scale architectures, which require significant computational resources and limit deployment on resource-constrained devices. In this paper, we propose a simple yet effective token pruning method for spiking transformers, termed TP-Spikformer, that reduces storage and computational overhead while maintaining competitive performance. Specifically, we first introduce a heuristic spatiotemporal information-retaining criterion that comprehensively evaluates tokens' importance, assigning higher scores to informative tokens for retention and lower scores to uninformative ones for pruning. Based on this criterion, we propose an information-retaining token pruning framework that employs a block-level early stopping strategy for uninformative tokens, instead of removing them outright. This also helps preserve more information during token pruning. We demonstrate the effectiveness, efficiency and scalability of TP-Spikformer through extensive experiments across diverse architectures, including Spikformer, QKFormer and Spike-driven Transformer V1 and V3, and a range of tasks such as image classification, object detection, semantic segmentation and event-based object tracking. Particularly, TP-Spikformer performs well in a training-free manner. These results reveal its potential as an efficient and practical solution for deploying SNNs in real-world applications with limited computational resources.
Authors: Swapnil Parekh
Abstract: Image captioning models are encoder-decoder architectures trained on large-scale image-text datasets, making them susceptible to adversarial attacks. We present CaptionFool, a novel universal (input-agnostic) adversarial attack against state-of-the-art transformer-based captioning models. By modifying only 7 out of 577 image patches (approximately 1.2% of the image), our attack achieves 94-96% success rate in generating arbitrary target captions, including offensive content. We further demonstrate that CaptionFool can generate "slang" terms specifically designed to evade existing content moderation filters. Our findings expose critical vulnerabilities in deployed vision-language models and underscore the urgent need for robust defenses against such attacks. Warning: This paper contains model outputs which are offensive in nature.
Authors: Xianhao Zhou, Jianghao Wu, Lanfeng Zhong, Ku Zhao, Jinlong He, Shaoting Zhang, Guotai Wang
Abstract: Cone-beam CT (CBCT) is routinely acquired in radiotherapy but suffers from severe artifacts and unreliable Hounsfield Unit (HU) values, limiting its direct use for dose calculation. Synthetic CT (sCT) generation from CBCT is therefore an important task, yet paired CBCT--CT data are often unavailable or unreliable due to temporal gaps, anatomical variation, and registration errors. In this work, we introduce rectified flow (RF) into unpaired CBCT-to-CT translation in medical imaging. Although RF is theoretically compatible with unpaired learning through distribution-level coupling and deterministic transport, its practical effectiveness under small medical datasets and limited batch sizes remains underexplored. Direct application with random or batch-local pseudo pairing can produce unstable supervision due to semantically mismatched endpoint samples. To address this challenge, we propose Retrieval-Augmented Flow Matching (RAFM), which adapts RF to the medical setting by constructing retrieval-guided pseudo pairs using a frozen DINOv3 encoder and a global CT memory bank. This strategy improves empirical coupling quality and stabilizes unpaired flow-based training. Experiments on SynthRAD2023 under a strict subject-level true-unpaired protocol show that RAFM outperforms existing methods across FID, MAE, SSIM, PSNR, and SegScore. The code is available at https://github.com/HiLab-git/RAFM.git.
Authors: Yafei Zhang, Shuaitian Song, Huafeng Li, Shujuan Wang, Yu Liu
Abstract: In real-world vision systems,haze removal is required not only to enhance image visibility but also to meet the specific needs of diverse downstream tasks.To address this challenge,we propose a novel adaptive dynamic dehazing framework that incorporates a closed-loop optimization mechanism.It enables feedback-driven refinement based on downstream task performance and user instruction-guided adjustment during inference,allowing the model to satisfy the specific requirements of multiple downstream tasks without retraining.Technically,our framework integrates two complementary and innovative mechanisms: (1)a task feedback loop that dynamically modulates dehazing outputs based on performance across multiple downstream tasks,and (2) a text instruction interface that allows users to specify high-level task preferences.This dual-guidance strategy enables the model to adapt its dehazing behavior after training,tailoring outputs in real time to the evolving needs of multiple tasks.Extensive experiments across various vision tasks demonstrate the strong effectiveness,robustness,and generalizability of our approach.These results establish a new paradigm for interactive,task-adaptive dehazing that actively collaborates with downstream applications.
Authors: Ke Cao, Xuanhua He, Xueheng Li, Lingting Zhu, Yingying Wang, Ao Ma, Zhanjie Zhang, Man Zhou, Chengjun Xie, Jie Zhang
Abstract: Pansharpening aims to generate high-resolution multi-spectral images by fusing the spatial detail of panchromatic images with the spectral richness of low-resolution MS data. However, most existing methods are evaluated under limited, low-resolution settings, limiting their generalization to real-world, high-resolution scenarios. To bridge this gap, we systematically investigate the data, algorithmic, and computational challenges of cross-scale pansharpening. We first introduce PanScale, the first large-scale, cross-scale pansharpening dataset, accompanied by PanScale-Bench, a comprehensive benchmark for evaluating generalization across varying resolutions and scales. To realize scale generalization, we propose ScaleFormer, a novel architecture designed for multi-scale pansharpening. ScaleFormer reframes generalization across image resolutions as generalization across sequence lengths: it tokenizes images into patch sequences of the same resolution but variable length proportional to image scale. A Scale-Aware Patchify module enables training for such variations from fixed-size crops. ScaleFormer then decouples intra-patch spatial feature learning from inter-patch sequential dependency modeling, incorporating Rotary Positional Encoding to enhance extrapolation to unseen scales. Extensive experiments show that our approach outperforms SOTA methods in fusion quality and cross-scale generalization. The datasets and source code are available upon acceptance.
Authors: Juan A. Castro-Silva, Maria N. Moreno Garcia, Diego H. Peluffo-Ordo\~nez
Abstract: The current methods for diagnosing Alzheimer Disease using Magnetic Resonance Imaging (MRI) have significant limitations. Many previous studies used 2D Transformers to analyze individual brain slices independently, potentially losing critical 3D contextual information. Region of interest-based models often focus on only a few brain regions despite Alzheimer's affecting multiple areas. Additionally, most classification models rely on a single test, whereas diagnosing Alzheimer's requires a multifaceted approach integrating diverse data sources for a more accurate assessment. This study introduces a novel methodology called the Multiple Inputs and Mixed Data 3D Vision Transformer (MIMD-3DVT). This method processes consecutive slices together to capture the feature dimensions and spatial information, fuses multiple 3D ROI imaging data inputs, and integrates mixed data from demographic factors, cognitive assessments, and brain imaging. The proposed methodology was experimentally evaluated using a combined dataset that included the Alzheimer's Disease Neuroimaging Initiative (ADNI), the Australian Imaging, Biomarker, and Lifestyle Flagship Study of Ageing (AIBL), and the Open Access Series of Imaging Studies (OASIS). Our MIMD-3DVT, utilizing single or multiple ROIs, achieved an accuracy of 97.14%, outperforming the state-of-the-art methods in distinguishing between Normal Cognition and Alzheimer's Disease.
Authors: Yu Wang, Shengjie Zhao
Abstract: Weakly supervised video anomaly detection (WS-VAD) involves identifying the temporal intervals that contain anomalous events in untrimmed videos, where only video-level annotations are provided as supervisory signals. However, a key limitation persists in WS-VAD, as dense frame-level annotations are absent, which often leaves existing methods struggling to learn anomaly semantics effectively. To address this issue, we propose a novel framework named LAS-VAD, short for Learning Anomaly Semantics for WS-VAD, which integrates anomaly-connected component mechanism and intention awareness mechanism. The former is designed to assign video frames into distinct semantic groups within a video, and frame segments within the same group are deemed to share identical semantic information. The latter leverages an intention-aware strategy to distinguish between similar normal and abnormal behaviors (e.g., taking items and stealing). To further model the semantic information of anomalies, as anomaly occurrence is accompanied by distinct characteristic attributes (i.e., explosions are characterized by flames and thick smoke), we additionally incorporate anomaly attribute information to guide accurate detection. Extensive experiments on two benchmark datasets, XD-Violence and UCF-Crime, demonstrate that our LAS-VAD outperforms current state-of-the-art methods with remarkable gains.
Authors: Yihua Shao, Kang Chen, Feng Xue, Siyu Chen, Long Bai, Hongyuan Yu, Hao Tang, Jinlin Wu, Nassir Navab
Abstract: In operating rooms (OR), world-scale multi-view 3D tracking supports downstream applications such as surgeon behavior recognition, where physically meaningful quantities such as distances and motion statistics must be measured in meters. However, real clinical deployments rarely satisfy the geometric prerequisites for stable multi-view fusion and tracking: camera calibration and RGB-D registration are always unreliable, leading to cross-view geometric inconsistency that produces "ghosting" during fusion and degrades 3D trajectories in a shared OR coordinate frame. To address this, we introduce Geometry OR Tracker, a two-stage pipeline that first rectifies imprecise calibration into a scaleconsistent and geometrically consistent camera setup with a single global scale via a Multi-view Metric Geometry Rectification module, and then performs Occlusion-Robust 3D Point Tracking directly in the unified OR world frame. On the MM-OR benchmark, improved geometric consistency translates into tracking gains: our rectification front-end reduces cross-view depth disagreement by more than 30$\times$ compared to raw calibration. Ablation studies further demonstrate the relationship between calibration quality and tracking accuracy, showing that improved geometric consistency yields stronger world-frame tracking.
Authors: Yilian Liu, Xiaojun Jia, Guoshun Nan, Jiuyang Lyu, Zhican Chen, Tao Guan, Shuyuan Luo, Zhongyi Zhai, Yang Liu
Abstract: Multimodal Large Language Models (MLLMs) have achieved remarkable performance but remain vulnerable to jailbreak attacks that can induce harmful content and undermine their secure deployment. Previous studies have shown that introducing additional inference steps, which disrupt security attention, can make MLLMs more susceptible to being misled into generating malicious content. However, these methods rely on single-image masking or isolated visual cues, which only modestly extend reasoning paths and thus achieve limited effectiveness, particularly against strongly aligned commercial closed-source models. To address this problem, in this paper, we propose Multi-Image Dispersion and Semantic Reconstruction (MIDAS), a multimodal jailbreak framework that decomposes harmful semantics into risk-bearing subunits, disperses them across multiple visual clues, and leverages cross-image reasoning to gradually reconstruct the malicious intent, thereby bypassing existing safety mechanisms. The proposed MIDAS enforces longer and more structured multi-image chained reasoning, substantially increases the model's reliance on visual cues while delaying the exposure of malicious semantics and significantly reducing the model's security attention, thereby improving the performance of jailbreak against advanced MLLMs. Extensive experiments across different datasets and MLLMs demonstrate that the proposed MIDAS outperforms state-of-the-art jailbreak attacks for MLLMs and achieves an average attack success rate of 81.46% across 4 closed-source MLLMs. Our code is available at this [link](https://github.com/Winnie-Lian/MIDAS).
Authors: Yongbo He, Zirun Guo, Tao Jin
Abstract: Adapting pretrained multi-modal models to evolving test-time distributions, known as multi-modal test-time adaptation, presents a significant challenge. Existing methods frequently encounter negative transfer in the unbiased modality and catastrophic forgetting in the biased modality. To address these challenges, we propose Decoupling Adaptation for Stability and Plasticity (DASP), a novel diagnose-then-mitigate framework. Our analysis reveals a critical discrepancy within the unified latent space: the biased modality exhibits substantially higher interdimensional redundancy (i.e., strong correlations across feature dimensions) compared to the unbiased modality. Leveraging this insight, DASP identifies the biased modality and implements an asymmetric adaptation strategy. This strategy employs a decoupled architecture where each modality-specific adapter is divided into stable and plastic components. The asymmetric mechanism works as follows: for the biased modality, which requires plasticity, the plastic component is activated and updated to capture domain-specific information, while the stable component remains fixed. Conversely, for the unbiased modality, which requires stability, the plastic component is bypassed, and the stable component is updated using KL regularization to prevent negative transfer. This asymmetric design enables the model to adapt flexibly to new domains while preserving generalizable knowledge. Comprehensive evaluations on diverse multi-modal benchmarks demonstrate that DASP significantly outperforms state-of-the-art methods.
Authors: Qin Guo, Tianyu Yang, Xuanhua He, Fei Shen, Yong Zhang, Zhuoliang Kang, Xiaoming Wei, Dan Xu
Abstract: Production-ready human video generation requires digital actors to maintain strictly consistent full-body identities across dynamic shots, viewpoints and motions, a setting that remains challenging for existing methods. Prior methods often suffer from face-centric behavior that neglects body-level consistency, or produce copy-paste artifacts where subjects appear rigid due to pose locking. We present Actor-18M, a large-scale human video dataset designed to capture identity consistency under unconstrained viewpoints and environments. Actor-18M comprises 1.6M videos with 18M corresponding human images, covering both arbitrary views and canonical three-view representations. Leveraging Actor-18M, we propose WildActor, a framework for any-view conditioned human video generation. We introduce an Asymmetric Identity-Preserving Attention mechanism coupled with a Viewpoint-Adaptive Monte Carlo Sampling strategy that iteratively re-weights reference conditions by marginal utility for balanced manifold coverage. Evaluated on the proposed Actor-Bench, WildActor consistently preserves body identity under diverse shot compositions, large viewpoint transitions, and substantial motions, surpassing existing methods in these challenging settings.
Authors: Cencen Liu (University of Electronic Science and Technology of China), Dongyang Zhang (University of Electronic Science and Technology of China, Ubiquitous Intelligence and Trusted Services Key Laboratory of Sichuan Province), Wen Yin (University of Electronic Science and Technology of China), Jielei Wang (University of Electronic Science and Technology of China, Ubiquitous Intelligence and Trusted Services Key Laboratory of Sichuan Province), Tianyu Li (University of Electronic Science and Technology of China), Ji Guo (University of Electronic Science and Technology of China), Wenbo Jiang (University of Electronic Science and Technology of China), Guoqing Wang (University of Electronic Science and Technology of China), Guoming Lu (University of Electronic Science and Technology of China, Ubiquitous Intelligence and Trusted Services Key Laboratory of Sichuan Province)
Abstract: Visual autoregressive (VAR) models have recently emerged as a promising alternative for image generation, offering stable training, non-iterative inference, and high-fidelity synthesis through next-scale prediction. This encourages the exploration of VAR for image super-resolution (ISR), yet its application remains underexplored and faces two critical challenges: locality-biased attention, which fragments spatial structures, and residual-only supervision, which accumulates errors across scales, severely compromises global consistency of reconstructed images. To address these issues, we propose AlignVAR, a globally consistent visual autoregressive framework tailored for ISR, featuring two key components: (1) Spatial Consistency Autoregression (SCA), which applies an adaptive mask to reweight attention toward structurally correlated regions, thereby mitigating excessive locality and enhancing long-range dependencies; and (2) Hierarchical Consistency Constraint (HCC), which augments residual learning with full reconstruction supervision at each scale, exposing accumulated deviations early and stabilizing the coarse-to-fine refinement process. Extensive experiments demonstrate that AlignVAR consistently enhances structural coherence and perceptual fidelity over existing generative methods, while delivering over 10x faster inference with nearly 50% fewer parameters than leading diffusion-based approaches, establishing a new paradigm for efficient ISR.
Authors: Chenggang Rong, Tao Han, Zhiyuan Zhao, Yaowu Fan, Jia Wan, Song Guo, Yuan Yuan, Junyu Gao
Abstract: Counting is a core capability for multimodal large language models (MLLMs), yet there is no unified counting dataset to rigorously evaluate this ability across image, text, and audio. We present UNICBench, a unified multimodal, multi level counting benchmark and evaluation toolkit with accurate ground truth, deterministic numeric parsing, and stratified reporting. The corpus comprises 5,300 images (5,508 QA), 872 documents (5,888 QA), and 2,069 audio clips (2,905 QA), annotated with a three level capability taxonomy and difficulty tags. Under a standardized protocol with fixed splits/prompts/seeds and modality specific matching rules, we evaluate 45 state-of-the-art MLLMs across modalities. Results show strong performance on some basic counting tasks but significant gaps on reasoning and the hardest partitions, highlighting long-tail errors and substantial headroom for improving general counting. UNICBench offers a rigorous and comparable basis for measurement and a public toolkit to accelerate progress.
Authors: Keiller Nogueira, Codrut-Andrei Diaconu, D\'avid Kerekes, Jakob Gawlikowski, C\'edric L\'eonard, Nassim Ait Ali Braham, June Moh Goo, Zichao Zeng, Zhipeng Liu, Pallavi Jain, Andrea Nascetti, Ronny H\"ansch
Abstract: High-quality pixel-level annotations are essential for the semantic segmentation of remote sensing imagery. However, such labels are expensive to obtain and often affected by noise due to the labor-intensive and time-consuming nature of pixel-wise annotation, which makes it challenging for human annotators to label every pixel accurately. Annotation errors can significantly degrade the performance and robustness of modern segmentation models, motivating the need for reliable mechanisms to identify and quantify noisy training samples. This paper introduces a novel Data-Centric benchmark, together with a novel, publicly available dataset and two techniques for identifying, quantifying, and ranking training samples according to their level of label noise in remote sensing semantic segmentation. Such proposed methods leverage complementary strategies based on model uncertainty, prediction consistency, and representation analysis, and consistently outperform established baselines across a range of experimental settings. The outcomes of this work are publicly available at https://github.com/keillernogueira/label_noise_segmentation.
URLs: https://github.com/keillernogueira/label_noise_segmentation.
Authors: Honghao Cai, Xiangyuan Wang, Yunhao Bai, Tianze Zhou, Sijie Xu, Yuyang Hao, Zezhou Cui, Yuyuan Yang, Wei Zhu, Yibo Chen, Xu Tang, Yao Hu, Zhen Li
Abstract: Multi-subject image generation requires seamlessly harmonizing multiple reference identities within a coherent scene. However, existing methods relying on rigid spatial masks or localized attention often struggle with the "stability-plasticity dilemma," particularly failing in tasks that require complex structural deformations, such as identity-preserving age transformation. To address this, we present IdGlow, a mask-free, progressive two-stage framework built upon Flow Matching diffusion models. In the supervised fine-tuning (SFT) stage, we introduce task-adaptive timestep scheduling aligned with diffusion generative dynamics: a linear decay schedule that progressively relaxes constraints for natural group composition, and a temporal gating mechanism that concentrates identity injection within a critical semantic window, successfully preserving adult facial semantics without overriding child-like anatomical structures. To resolve attribute leakage and semantic ambiguity without explicit layout inputs, we further integrate a badcase-driven Vision-Language Model (VLM) for precise, context-aware prompt synthesis. In the second stage, we design a Fine-Grained Group-Level Direct Preference Optimization (DPO) with a weighted margin formulation to simultaneously eliminate multi-subject artifacts, elevate texture harmony, and recalibrate identity fidelity towards real-world distributions. Extensive experiments on two challenging benchmarks -- direct multi-person fusion and age-transformed group generation -- demonstrate that IdGlow fundamentally mitigates the stability-plasticity conflict, achieving a superior Pareto balance between state-of-the-art facial fidelity and commercial-grade aesthetic quality.
Authors: Changxing Liu, Zichen Chao, Siheng Chen
Abstract: Collaborative perception leverages data exchange among multiple agents to enhance overall perception capabilities. However, heterogeneity across agents introduces domain gaps that hinder collaboration, and this is further exacerbated by an underexplored issue: modality isolation. It arises when multiple agents with different modalities never co-occur in any training data frame, enlarging cross-modal domain gaps. Existing alignment methods rely on supervision from spatially overlapping observations, thus fail to handle modality isolation. To address this challenge, we propose CodeAlign, the first efficient, co-occurrence-free alignment framework that smoothly aligns modalities via cross-modal feature-code-feature(FCF) translation. The key idea is to explicitly identify the representation consistency through codebook, and directly learn mappings between modality-specific feature spaces, thereby eliminating the need for spatial correspondence. Codebooks regularize feature spaces into code spaces, providing compact yet expressive representations. With a prepared code space for each modality, CodeAlign learns FCF translations that map features to the corresponding codes of other modalities, which are then decoded back into features in the target code space, enabling effective alignment. Experiments show that, when integrating three modalities, CodeAlign requires only 8% of the training parameters of prior alignment methods, reduces communication load by 1024x, and achieves state-of-the-art perception performance on both OPV2V and DAIR-V2X dataset. Code will be released on https://github.com/cxliu0314/CodeAlign.
Authors: Lijing Cai, Zhan Shi, Chenglong Huang, Jinyao Wu, Qiping Li, Zikang Huo, Linsen Chen, Chongde Zi, Xun Cao
Abstract: Recently, Spectral Compressive Imaging (SCI) has achieved remarkable success, unlocking significant potential for dynamic spectral vision. However, existing reconstruction methods, primarily image-based, suffer from two limitations: (i) Encoding process masks spatial-spectral features, leading to uncertainty in reconstructing missing information from single compressed measurements, and (ii) The frame-by-frame reconstruction paradigm fails to ensure temporal consistency, which is crucial in the video perception. To address these challenges, this paper seeks to advance spectral reconstruction from the image level to the video level, leveraging the complementary features and temporal continuity across adjacent frames in dynamic scenes. Initially, we construct the first high-quality dynamic hyperspectral image dataset (DynaSpec), comprising 30 sequences obtained through frame-scanning acquisition. Subsequently, we propose the Propagation-Guided Spectral Video Reconstruction Transformer (PG-SVRT), which employs a spatial-then-temporal attention to effectively reconstruct spectral features from abundant video information, while using a bridged token to reduce computational complexity. Finally, we conduct simulation experiments to assess the performance of four SCI systems, and construct a DD-CASSI prototype for real-world data collection and benchmarking. Extensive experiments demonstrate that PG-SVRT achieves superior performance in reconstruction quality, spectral fidelity, and temporal consistency, while maintaining minimal FLOPs. Project page: https://github.com/nju-cite/DynaSpec
Authors: Jinfan Hu, Fanghua Yu, Zhiyuan You, Xiang Yin, Hongyu An, Xinqi Lin, Chao Dong, Jinjin Gu
Abstract: This position paper argues that the evaluation of modern visual processing systems should no longer be driven primarily by single-metric image quality assessment benchmarks, particularly in the era of generative and perception-oriented methods. Image restoration exemplifies this divergence: while objective IQA metrics enable reproducible, scalable evaluation, they have increasingly drifted apart from human perception and user preferences. We contend that this mismatch risks constraining innovation and misguiding research progress across visual processing tasks. Rather than rejecting metrics altogether, this paper calls for a rebalancing of evaluation paradigms, advocating a more human-centered, context-aware, and fine-grained approach to assessing the visual models' outcomes.
Authors: Xiaohan Zhao, Xinyi Shang, Jiacheng Liu, Zhiqiang Shen
Abstract: Dataset pruning has been widely studied for 2D images to remove redundancy and accelerate training, while particular pruning methods for 3D data remain largely unexplored. In this work, we study dataset pruning for 3D data, where its observed common long-tail class distribution nature make optimization under conventional evaluation metrics Overall Accuracy (OA) and Mean Accuracy (mAcc) inherently conflicting, and further make pruning particularly challenging. To address this, we formulate pruning as approximating the full-data expected risk with a weighted subset, which reveals two key errors: coverage error from insufficient representativeness and prior-mismatch bias from inconsistency between subset-induced class weights and target metrics. We propose representation-aware subset selection with per-class retention quotas for long-tail coverage, and prior-invariant teacher supervision using calibrated soft labels and embedding-geometry distillation. The retention quota also serves as a switch to control the OA-mAcc trade-off. Extensive experiments on 3D datasets show that our method can improve both metrics across multiple settings while adapting to different downstream preferences. Our code is available at https://github.com/XiaohanZhao123/3D-Dataset-Pruning.
Authors: Xiaokai Bai, Lianqing Zheng, Runwei Guan, Siyuan Cao, Huiliang Shen
Abstract: Collaborative perception (CP) enhances scene understanding through multi-agent information sharing. While LiDAR-centric systems offer precise geometry, high costs and performance degradation in adverse weather necessitate multi-modal alternatives. Despite dense visual semantics and robust spatial measurements, the synergy between cameras and 4D radar remains underexplored in collaborative settings. This work introduces RC-GeoCP, the first framework to explore the fusion of 4D radar and images in CP. To resolve misalignment caused by depth ambiguity and spatial dispersion across agents, RC-GeoCP establishes a radar-anchored geometric consensus. Specifically, Geometric Structure Rectification (GSR) aligns visual semantics with geometry derived from radar to generate spatially grounded, geometry-consistent representations. Uncertainty-Aware Communication (UAC) formulates selective transmission as a conditional entropy reduction process to prioritize informative features based on inter-agent disagreement. Finally, the Consensus-Driven Assembler (CDA) aggregates multi-agent information via shared geometric anchors to form a globally coherent representation. We establish the first unified radar-camera CP benchmark on V2X-Radar and V2X-R, demonstrating state-of-the-art performance with significantly reduced communication overhead. Code will be released soon.
Authors: Ying Liu, Yudong Han, Kean Shi, Liyuan Pan
Abstract: Recent multimodal large language models (MLLMs) widely adopt multi-layer visual feature fusion to enhance visual representation. However, existing approaches typically perform static concatenation or weighted aggregation after visual encoding, without intervening in the representation formation process itself. As a result, fine-grained details from early layers may be progressively suppressed during hierarchical abstraction. Moreover, directly introducing shallow-layer features into the language model often leads to semantic distribution mismatch with the visual feature space that the LLM's cross-attention layers were pretrained on, which typically requires additional adaptation or fine-tuning of the LLM. To address these limitations, we revisit visual representation learning from the perspective of representation evolution control and propose a cross-layer memory-modulated vision framework(SCVM). Specifically, we introduce a recursively updated cross-layer memory state inside the vision encoder to model long-range inter-layer dependencies. We further design a layer-wise feedback modulation mechanism that refreshes token representations at each layer based on the accumulated memory, thereby structurally regulating the representation evolution trajectory. In addition, we incorporate an auxiliary semantic alignment objective that explicitly supervises the final memory state, encouraging progressive compression and reinforcement of task-relevant information. Experimental results on multiple visual question answering and hallucination evaluation benchmarks demonstrate that SCVM achieves consistent performance improvements without expanding visual tokens, introducing additional vision encoders, or modifying or fine-tuning the language model.
Authors: Wentao Huang, Weimin Lyu, Peiliang Lou, Qingqiao Hu, Xiaoling Hu, Shahira Abousamra, Wenchao Han, Ruifeng Guo, Jiawei Zhou, Chao Chen, Chen Wang
Abstract: Computational pathology has advanced rapidly in recent years, driven by domain-specific image encoders and growing interest in using vision-language models to answer natural-language questions about diseases. Yet, the core problem behind pathology question-answering remains unsolved, considering that a gigapixel slide contains far more information than necessary for a given question. Pathologists naturally navigate tissue and morphology complexity by scanning broadly, and zooming in selectively according to the clinical questions. Current models, in contrast, rely on uniform patch sampling or broad attention maps, often attending equally to irrelevant regions while overlooking key visual evidence. In this work, we try to bring models closer to how humans actually examine slides. We propose a question-guided, tissue-aware, and coarse-to-fine retrieval framework, HistoSelect, that consists of two key components: a group sampler that identifies question-relevant tissue regions, followed by a patch selector that retrieves the most informative patches within those regions. By selecting only the most informative patches, our method becomes significantly more efficient: reducing visual token usage by 70% on average, while improving accuracy across three pathology QA tasks. Evaluated on 356,000 question-answer pairs, our approach outperforms existing methods and produces answers grounded in interpretable, pathologist-consistent regions. Our results suggest that bringing human-like search and attention patterns into WSI reasoning is a promising direction for building practical and reliable pathology VLMs.
Authors: Daniel Tweneboah Anyimadu, Mohammed M. Abdelsamea, Ahmed Karam Eldaly
Abstract: Low-field magnetic resonance imaging (MRI) provides affordable access to diagnostic imaging but suffers from prolonged acquisition and limited image quality. Accelerated imaging can be achieved with k-space undersampling, while super-resolution (SR) and image quality transfer (IQT) methods typically rely on spatial-domain post-processing. In this work, we propose a novel framework for reconstructing high-field MR like images directly from undersampled low-field k-space. Our approach employs a k-space dual channel U-Net that processes the real and imaginary components of undersampled k-space to restore missing frequency content. Experiments on low-field brain MRI demonstrate that our k-space-driven image enhancement consistently outperforms the counterpart spatial-domain method. Furthermore, reconstructions from undersampled k-space achieve image quality comparable to full k-space acquisitions. To the best of our knowledge, this is the first work that investigates low-field MRI SR/IQT directly from undersampled k-space.
Authors: Youngjin Yoo, Han Liu, Bogdan Georgescu, Yanbo Zhang, Sasa Grbic, Michael Baumgartner, Thomas J. Re, Jyotipriya Das, Poikavila Ullaskrishnan, Eva Eibenberger, Andrei Chekkoury, Uttam K. Bodanapally, Savvas Nicolaou, Pina C. Sanelli, Thomas J. Schroeppel, Yvonne W. Lui, Eli Gibson
Abstract: Foundation models pre-trained on large-scale datasets demonstrate strong transfer learning capabilities; however, their adaptation to complex multi-label diagnostic tasks-such as comprehensive head CT finding detection-remains understudied. Standard parameter-efficient fine-tuning methods such as LoRA apply uniform adaptations across pathology types, which may limit performance for diverse medical findings. We propose a Mixture of Low-Rank Experts (MoLRE) framework that extends LoRA with multiple specialized low-rank adapters and unsupervised soft routing. This approach enables conditional feature adaptation with less than 0.5% additional parameters and without explicit pathology supervision. We present a comprehensive benchmark of MoLRE across six state-of-the-art medical imaging foundation models spanning 2D and 3D architectures, general-domain, medical-domain, and head CT-specific pretraining, and model sizes ranging from 7M to 431M parameters. Using over 70,000 non-contrast head CT scans with 75 annotated findings-including hemorrhage, infarction, trauma, mass lesions, structural abnormalities, and chronic changes-our experiments demonstrate consistent performance improvements across all models. Gains vary substantially: general-purpose and medical-domain models show the largest improvements (DINOv3-Base: +4.6%; MedGemma: +4.3%), whereas 3D CT-specialized or very large models show more modest gains (+0.2-1.3%). The combination of MoLRE and MedGemma achieves the highest average detection AUC of 0.917. These findings highlight the importance of systematic benchmarking on target clinical tasks, as pretraining domain, architecture, and model scale interact in non-obvious ways.
Authors: Yushan Han, Hui Zhang, Qiming Xia, Yi Jin, Yidong Li
Abstract: Collaborative perception empowers autonomous agents to share complementary information and overcome perception limitations. While early fusion offers more perceptual complementarity and is inherently robust to model heterogeneity, its high communication cost has limited its practical deployment, prompting most existing works to favor intermediate or late fusion. To address this, we propose a communication-efficient early Collaborative perception framework that incorporates LiDAR Completion to restore scene completeness under sparse transmission, dubbed as CoLC. Specifically, the CoLC integrates three complementary designs. First, each neighbor agent applies Foreground-Aware Point Sampling (FAPS) to selectively transmit informative points that retain essential structural and contextual cues under bandwidth constraints. The ego agent then employs Completion-Enhanced Early Fusion (CEEF) to reconstruct dense pillars from the received sparse inputs and adaptively fuse them with its own observations, thereby restoring spatial completeness. Finally, the Dense-Guided Dual Alignment (DGDA) strategy enforces semantic and geometric consistency between the enhanced and dense pillars during training, ensuring consistent and robust feature learning. Experiments on both simulated and real-world datasets demonstrate that CoLC achieves superior perception-communication trade-offs and remains robust under heterogeneous model settings. The code is available at https://github.com/CatOneTwo/CoLC.
Authors: Guoquan Wei, Liu Shi, Shaoyu Wang, Mohan Li, Cunfeng Wei, Qiegen Liu
Abstract: Noise and artifacts during computed tomography (CT) scans are a fundamental challenge affecting disease diagnosis. However, current methods either involve excessively long reconstruction times or rely on data-driven models for optimization, failing to adequately consider the valuable information inherent in the data itself, especially medical 3D data. This work proposes a reconstruction method under ultra-low raw data conditions, requiring no external data and avoiding lengthy pre-training processes. By leveraging spatial nonlocal similarity and the conjugate properties of the projection domain to generate pseudo-3D data for self-supervised training, high-fidelity results can be achieved in a very short time. Extensive experiments demonstrate that this method not only mitigates detector-induced ring artifacts but also exhibits unprecedented capabilities in detail recovery. This method provides a new paradigm for research using unlabeled raw projection data. Code is available at https://github.com/yqx7150/SCOUT.
Authors: Xingguo Xu, Zhanyu Liu, Weixiang Zhou, Yuansheng Gao, Junjie Cao, Yuhao Wang, Jixiang Luo, Dell Zhang
Abstract: Multi-modal object Re-Identification (ReID) aims to exploit complementary information from different modalities to retrieve specific objects. However, existing methods often rely on hard token filtering or simple fusion strategies, which can lead to the loss of discriminative cues and increased background interference. To address these challenges, we propose STMI, a novel multi-modal learning framework consisting of three key components: (1) Segmentation-Guided Feature Modulation (SFM) module leverages SAM-generated masks to enhance foreground representations and suppress background noise through learnable attention modulation; (2) Semantic Token Reallocation (STR) module employs learnable query tokens and an adaptive reallocation mechanism to extract compact and informative representations without discarding any tokens; (3) Cross-Modal Hypergraph Interaction (CHI) module constructs a unified hypergraph across modalities to capture high-order semantic relationships. Extensive experiments on public benchmarks (i.e., RGBNT201, RGBNT100, and MSVR310) demonstrate the effectiveness and robustness of our proposed STMI framework in multi-modal ReID scenarios.
Authors: Yihui Li, Chengxin Lv, Zichen Tang, Hongyu Yang, Di Huang
Abstract: We present TokenSplat, a feed-forward framework for joint 3D Gaussian reconstruction and camera pose estimation from unposed multi-view images. At its core, TokenSplat introduces a Token-aligned Gaussian Prediction module that aligns semantically corresponding information across views directly in the feature space. Guided by coarse token positions and fusion confidence, it aggregates multi-scale contextual features to enable long-range cross-view reasoning and reduce redundancy from overlapping Gaussians. To further enhance pose robustness and disentangle viewpoint cues from scene semantics, TokenSplat employs learnable camera tokens and an Asymmetric Dual-Flow Decoder (ADF-Decoder) that enforces directionally constrained communication between camera and image tokens. This maintains clean factorization within a feed-forward architecture, enabling coherent reconstruction and stable pose estimation without iterative refinement. Extensive experiments demonstrate that TokenSplat achieves higher reconstruction fidelity and novel-view synthesis quality in pose-free settings, and significantly improves pose estimation accuracy compared to prior pose-free methods. Project page: https://kidleyh.github.io/tokensplat/.
Authors: Marry Kong, Rina Buoy, Sovisal Chenda, Nguonly Taing, Masakazu Iwamura, Koichi Kise
Abstract: Khmer is a low-resource language characterized by a complex script, presenting significant challenges for optical character recognition (OCR). While document printed text recognition has advanced because of available datasets, performance on other modalities, such as handwritten and scene text, remains limited by data scarcity. Training modality-specific models for each modality does not allow cross-modality transfer learning, from which modalities with limited data could otherwise benefit. Moreover, deploying many modality-specific models results in significant memory overhead and requires error-prone routing each input image to the appropriate model. On the other hand, simply training on a combined dataset with a non-uniform data distribution across different modalities often leads to degraded performance on underrepresented modalities. To address these, we propose a universal Khmer text recognition (UKTR) framework capable of handling diverse text modalities. Central to our method is a novel modality-aware adaptive feature selection (MAFS) technique designed to adapt visual features according to a particular input image modality and enhance recognition robustness across modalities. Extensive experiments demonstrate that our model achieves state-of-the-art (SoTA) performance. Furthermore, we introduce the first comprehensive benchmark for universal Khmer text recognition, which we release to the community to facilitate future research. Our datasets and models can be accessible via this gated repository\footnote{in review}.
Authors: Marry Kong, Rina Buoy, Sovisal Chenda, Nguonly Taing, Masakazu Iwamura, Koichi Kise
Abstract: While document layout analysis for Latin scripts has advanced significantly, driven by the advent of large multimodal models (LMMs), progress for the Khmer language remains constrained because of the scarcity of annotated training data. This gap is particularly acute for scene documents, where perspective distortions and complex backgrounds challenge traditional methods. Given the structural complexities of Khmer script, such as diacritics and multi-layer character stacking, existing Latin-based layout analysis models fail to accurately delineate semantic layout units, particularly for dense text regions (e.g., list items). In this paper, we present the first comprehensive study on Khmer scene document layout detection. We contribute a novel framework comprising three key elements: (1) a robust training and benchmarking dataset specifically for Khmer scene layouts; (2) an open-source document augmentation tool capable of synthesizing realistic scene documents to scale training data; and (3) layout detection baselines utilizing YOLO-based architectures with oriented bounding boxes (OBB) to handle geometric distortions. To foster further research in the Khmer document analysis and recognition (DAR) community, we release our models, code, and datasets in this gated repository (in review).
Authors: Rui Ma, Yifeng Wang, Ziteng Yang, Xinghui Li
Abstract: Visual analysis and reconstruction of pipeline inner walls remain challenging in industrial inspection scenarios. This paper presents a dedicated reconstruction system for pipeline inner walls via industrial endoscopes, which is built on panoramic image stitching technology. Equipped with a custom graphical user interface (GUI), the system extracts key frames from endoscope video footage, and integrates polar coordinate transformation with image stitching techniques to unwrap annular video frames of pipeline inner walls into planar panoramic images. Experimental results demonstrate that the proposed method enables efficient processing of industrial endoscope videos, and the generated panoramic stitched images preserve all detailed features of pipeline inner walls in their entirety. This provides intuitive and accurate visual support for defect detection and condition assessment of pipeline inner walls. In comparison with the traditional frame-by-frame video review method, the proposed approach significantly elevates the efficiency of pipeline inner wall reconstruction and exhibits considerable engineering application value.
Authors: Qinghui He, Haifeng Zhang, Qiao Qin, Bo Liu, Xiuli Bi, Bin Xiao
Abstract: With the rapid advancement of generative models, generated image detection has become an important task in visual forensics. Although existing methods have achieved remarkable progress, they often rely, after training, on only a small subset of highly salient forgery cues, which limits their ability to generalize to unseen generative mechanisms. We argue that reliably generated image detection should not depend on a single decision path but should preserve multiple judgment perspectives, enabling the model to understand the differences between real and generated images from diverse viewpoints. Based on this idea, we propose an anti-feature-collapse learning framework that filters task-irrelevant components and suppresses excessive overlap among different forgery cues in the representation space, preventing discriminative information from collapsing into a few dominant feature directions. This design maintains diverse and complementary evidence within the model, reduces reliance on a small set of salient cues, and enhances robustness under unseen generative settings. Extensive experiments on multiple public benchmarks demonstrate that the proposed method significantly outperforms the state-of-the-art approaches in cross-model scenarios, achieving an accuracy improvement of 5.02% and exhibiting superior generalization and detection reliability. The source code is available at https://github.com/Yanmou-Hui/DoU.
Authors: Rafi Hassan Chowdhury, Naimul Haque, Kaniz Fatiha
Abstract: Handwritten character classification in the Bengali script is a significant challenge due to the complexity and variability of the characters. The models commonly used for classification are often computationally expensive and data-hungry, making them unsuitable for resource-limited languages such as Bengali. In this experiment, we propose a novel, efficient, and lightweight Vision Transformer model that effectively classifies Bengali handwritten basic characters and digits, addressing several shortcomings of traditional methods. The proposed solution utilizes a deep convolutional neural network (DCNN) in a more simplified manner compared to traditional DCNN architectures, with the aim of reducing computational burden. With only 0.65 million parameters, a model size of 0.62 MB, and 0.16 GFLOPs, our model, BornoViT, is significantly lighter than current state-of-the-art models, making it more suitable for resource-limited environments, which is essential for Bengali handwritten character classification. BornoViT was evaluated on the BanglaLekha Isolated dataset, achieving an accuracy of 95.77%, and demonstrating superior efficiency compared to existing state-of-the-art approaches. Furthermore, the model was evaluated on our self-collected dataset, Bornomala, consisting of approximately 222 samples from different age groups, where it achieved an accuracy of 91.51%.
Authors: Adam Marcus, Paul Bentley, Daniel Rueckert
Abstract: Stroke is a major cause of death and disability worldwide. Accurate outcome and evolution prediction has the potential to revolutionize stroke care by individualizing clinical decision-making leading to better outcomes. However, despite a plethora of attempts and the rich data provided by neuroimaging, modelling the ultimate fate of brain tissue remains a challenging task. In this work, we apply recent ideas in the field of diffusion probabilistic models to generate a self-supervised semantically meaningful stroke representation from Computed Tomography (CT) images. We then improve this representation by extending the method to accommodate longitudinal images and the time from stroke onset. The effectiveness of our approach is evaluated on a dataset consisting of 5,824 CT images from 3,573 patients across two medical centers with minimal labels. Comparative experiments show that our method achieves the best performance for predicting next-day severity and functional outcome at discharge.
Authors: Zhenyu Zhou, Defang Chen, Siwei Lyu, Chun Chen, Can Wang
Abstract: Text-to-image diffusion models have achieved unprecedented success but still struggle to produce high-quality results under limited sampling budgets. Existing training-free sampling acceleration methods are typically developed independently, leaving the overall performance and compatibility among these methods unexplored. In this paper, we bridge this gap by systematically elucidating the design space, and our comprehensive experiments identify the sampling time schedule as the most pivotal factor. Inspired by the geometric properties of diffusion models revealed through the Frenet-Serret formulas, we propose constant total rotation schedule (TORS), a scheduling strategy that ensures uniform geometric variation along the sampling trajectory. TORS outperforms previous training-free acceleration methods and produces high-quality images with 10 sampling steps on Flux.1-Dev and Stable Diffusion 3.5. Extensive experiments underscore the adaptability of our method to unseen models, hyperparameters, and downstream applications.
Authors: Zikang Xu, Ruinan Jin, Xiaoxiao Li
Abstract: Tool-using medical agents can improve chest X-ray question answering by orchestrating specialized vision and language modules, but this added pipeline complexity also creates new pathways for demographic bias beyond standalone models. We present ours (Decomposing Unfairness in Chest X-ray agents), a systematic audit of chest X-ray agents instantiated with MedRAX. To localize where disparities arise, we introduce a stage-wise fairness decomposition that separates end-to-end bias from three agent-specific sources: tool exposure bias (utility gaps conditioned on tool presence), tool transition bias (subgroup differences in tool-routing patterns), and model reasoning bias (subgroup differences in synthesis behaviors). Extensive experiments on tool-used based agentic frameworks across five driver backbones reveal that (i) demographic gaps persist in end-to-end performance, with equalized odds up to 20.79%, and the lowest fairness-utility tradeoff down to 28.65%, and (ii) intermediate behaviors, tool usage, transition patterns, and reasoning traces exhibit distinct subgroup disparities that are not predictable from end-to-end evaluation alone (e.g., conditioned on segmentation-tool availability, the subgroup utility gap reaches as high as 50%). Our findings underscore the need for process-level fairness auditing and debiasing to ensure the equitable deployment of clinical agentic systems. Code is available here: https://anonymous.4open.science/r/DUCK-E5FE/README.md
Authors: Ruiyu Yan, Hanqi Jiang, Yi Pan, Xiaobo Li, Tianming Liu, Xi Jiang, Lin Zhao
Abstract: We propose the Neural Functional Alignment Space (NFAS), a brain-referenced representational framework for characterizing artificial neural networks on equal functional grounds. NFAS departs from conventional alignment approaches that rely on layer-wise features or task-specific activations by modeling the intrinsic dynamical evolution of stimulus representations across network depth. Specifically, we model layer-wise embeddings as a depth-wise dynamical trajectory and apply Dynamic Mode Decomposition (DMD) to extract the stable mode. This representation is then projected into a biologically anchored coordinate system defined by distributed neural responses. We also introduce the Signal-to-Noise Consistency Index (SNCI) to quantify cross-model consistency at the modality level. Across 45 pretrained models spanning vision, audio, and language, NFAS reveals structured organization within this brain-referenced space, including modality-specific clustering and cross-modal convergence in integrative cortical systems. Our findings suggest that representation dynamics provide a principled basis for
Authors: Seemandhar Jain, Keshav Gupta, Kunal Gupta, Manmohan Chandraker
Abstract: The proliferation of neural radiance field (NeRF) research requires significant efforts to reimplement papers before building upon them. We introduce NERFIFY, a multi-agent framework that reliably converts NeRF research papers into trainable Nerfstudio plugins, in contrast to generic paper-to-code methods and frontier models like GPT-5 that usually fail to produce runnable code. NERFIFY achieves domain-specific executability through six key innovations: (1) Context-free grammar (CFG): LLM synthesis is constrained by Nerfstudio formalized as a CFG, ensuring generated code satisfies architectural invariants. (2) Graph-of-Thought code synthesis: Specialized multi-file-agents generate repositories in topological dependency order, validating contracts and errors at each node. (3) Compositional citation recovery: Agents automatically retrieve and integrate components (samplers, encoders, proposal networks) from citation graphs of references. (4) Visual feedback: Artifacts are diagnosed through PSNR-minima ROI analysis, cross-view geometric validation, and VLM-guided patching to iteratively improve quality. (5) Knowledge enhancement: Beyond reproduction, methods can be improved with novel optimizations. (6) Benchmarking: An evaluation framework is designed for NeRF paper-to-code synthesis across 30 diverse papers. On papers without public implementations, NERFIFY achieves visual quality matching expert human code (+/-0.5 dB PSNR, +/-0.2 SSIM) while reducing implementation time from weeks to minutes. NERFIFY demonstrates that a domain-aware design enables code translation for complex vision papers, potentiating accelerated and democratized reproducible research. Code, data and implementations will be publicly released.
Authors: Anmol Agarwal, Pranay Meshram, Sumer Singh, Saurav Suman, Andrew Lapp, Shahbuland Matiana, Louis Castricato, Spencer Frazier
Abstract: Recent advances in video generation have spurred the development of world models capable of simulating 3D-consistent environments and interactions with static objects. However, a significant limitation remains in their ability to model dynamic, reactive agents that can intelligently influence and interact with the world. To address this gap, we introduce COMBAT, a real-time, action-controlled world model trained on the complex 1v1 fighting game Tekken 3. Our work demonstrates that diffusion models can successfully simulate a dynamic opponent that reacts to player actions, learning its behavior implicitly. Our approach utilizes a 1.2 billion parameter Diffusion Transformer, conditioned on latent representations from a deep compression autoencoder. We employ state-of-the-art techniques, including causal distillation and diffusion forcing, to achieve real-time inference. Crucially, we observe the emergence of sophisticated agent behavior by training the model solely on single-player inputs, without any explicit supervision for the opponent's policy. Unlike traditional imitation learning methods, which require complete action labels, COMBAT learns effectively from partially observed data to generate responsive behaviors for a controllable Player 1. We present an extensive study and introduce novel evaluation methods to benchmark this emergent agent behavior, establishing a strong foundation for training interactive agents within diffusion-based world models.
Authors: Amir Belder, Ayellet Tal
Abstract: In recent years, various methods have been proposed for mesh analysis, each offering distinct advantages and often excelling on different object classes. We present a novel Mixture of Experts (MoE) framework designed to harness the complementary strengths of these diverse approaches. We propose a new gate architecture that encourages each expert to specialise in the classes it excels in. Our design is guided by two key ideas: (1) random walks over the mesh surface effectively capture the regions that individual experts attend to, and (2) an attention mechanism that enables the gate to focus on the areas most informative for each expert's decision-making. To further enhance performance, we introduce a dynamic loss balancing scheme that adjusts a trade-off between diversity and similarity losses throughout the training, where diversity prompts expert specialization, and similarity enables knowledge sharing among the experts. Our framework achieves state-of-the-art results in mesh classification, retrieval, and semantic segmentation tasks. Our code is available at: https://github.com/amirbelder/MME-Mixture-of-Mesh-Experts.
URLs: https://github.com/amirbelder/MME-Mixture-of-Mesh-Experts.
Authors: Cong Wang, Jinshan Pan, Liyan Wang, Wei Wang, Yang Yang
Abstract: We propose a simple yet effective UHDPromer, a neural discrimination-prompted Transformer, for Ultra-High-Definition (UHD) image restoration and enhancement. Our UHDPromer is inspired by an interesting observation that there implicitly exist neural differences between high-resolution and low-resolution features, and exploring such differences can facilitate low-resolution feature representation. To this end, we first introduce Neural Discrimination Priors (NDP) to measure the differences and then integrate NDP into the proposed Neural Discrimination-Prompted Attention (NDPA) and Neural Discrimination-Prompted Network (NDPN). The proposed NDPA re-formulates the attention by incorporating NDP to globally perceive useful discrimination information, while the NDPN explores a continuous gating mechanism guided by NDP to selectively permit the passage of beneficial content. To enhance the quality of restored images, we propose a super-resolution-guided reconstruction approach, which is guided by super-resolving low-resolution features to facilitate final UHD image restoration. Experiments show that UHDPromer achieves the best computational efficiency while still maintaining state-of-the-art performance on $3$ UHD image restoration and enhancement tasks, including low-light image enhancement, image dehazing, and image deblurring. The source codes and pre-trained models will be made available at https://github.com/supersupercong/uhdpromer.
Authors: Jie Li, Shengwei Tian, Long Yu, Xin Ning
Abstract: Existing point cloud completion methods struggle to balance high-quality reconstruction with computational efficiency. To address this, we propose PPC-MT, a novel parallel framework for point cloud completion leveraging a hybrid Mamba-Transformer architecture. Our approach introduces an innovative parallel completion strategy guided by Principal Component Analysis (PCA), which imposes a geometrically meaningful structure on unordered point clouds, transforming them into ordered sets and decomposing them into multiple subsets. These subsets are reconstructed in parallel using a multi-head reconstructor. This structured parallel synthesis paradigm significantly enhances the uniformity of point distribution and detail fidelity, while preserving computational efficiency. By integrating Mamba's linear complexity for efficient feature extraction during encoding with the Transformer's capability to model fine-grained multi-sequence relationships during decoding, PPC-MT effectively balances efficiency and reconstruction accuracy. Extensive quantitative and qualitative experiments on benchmark datasets, including PCN, ShapeNet-55/34, and KITTI, demonstrate that PPC-MT outperforms state-of-the-art methods across multiple metrics, validating the efficacy of our proposed framework.
Authors: Halil Ismail Helvaci, Justin Huber, Jihye Bae, Sen-ching Samson Cheung
Abstract: To empower the iterative assessments involved during a person's rehabilitation, automated assessment of a person's abilities during daily activities requires temporally precise segmentation of fine-grained actions in therapy videos. Existing temporal action segmentation (TAS) models struggle to capture sub-second micro-movements while retaining exercise context, blurring rapid phase transitions and limiting reliable downstream assessment of motor recovery. We introduce Multi-Membership Temporal Attention (MMTA), a high-resolution temporal transformer for fine-grained rehabilitation assessment. Unlike standard temporal attention, which assigns each frame a single attention context per layer, MMTA lets each frame attend to multiple locally normalized temporal attention windows within the same layer. We fuse these concurrent temporal views via feature-space overlap resolution, preserving competing local contexts near transitions while enabling longer-range reasoning through layer-wise propagation. This increases boundary sensitivity without additional depth or multi-stage refinement. MMTA supports both video and wearable IMU inputs within a unified single-stage architecture, making it applicable to both clinical and home settings. MMTA consistently improves over the Global Attention transformer, boosting Edit Score by +1.3 (Video) and +1.6 (IMU) on StrokeRehab while further improving 50Salads by +3.3. Ablations confirm that performance gains stem from multi-membership temporal views rather than architectural complexity, offering a practical solution for resource-constrained rehabilitation assessment.
Authors: Yu Luo, Guangyu Wei, Yangfan Li, Jieyu He, Yueming Lyu
Abstract: Segmentation of the main coronary artery from X-ray coronary angiography (XCA) sequences is crucial for the diagnosis of coronary artery diseases. However, this task is challenging due to issues such as blurred boundaries, inconsistent radiation contrast, complex motion patterns, and a lack of annotated images for training. Although Semi-Supervised Learning (SSL) can alleviate the annotation burden, conventional methods struggle with complicated temporal dynamics and unreliable uncertainty quantification. To address these challenges, we propose SAM3-based Teacher-student framework with Motion-Aware consistency and Progressive Confidence Regularization (SMART), a semi-supervised vessel segmentation approach for X-ray angiography videos. First, our method utilizes SAM3's unique promptable concept segmentation design and innovates a SAM3-based teacher-student framework to maximize the performance potential of both the teacher and the student. Second, we enhance segmentation by integrating the vessel mask warping technique and motion consistency loss to model complex vessel dynamics. To address the issue of unreliable teacher predictions caused by blurred boundaries and minimal contrast, we further propose a progressive confidence-aware consistency regularization to mitigate the risk of unreliable outputs. Extensive experiments on three datasets of XCA sequences from different institutions demonstrate that SMART achieves state-of-the-art performance while requiring significantly fewer annotations, making it particularly valuable for real-world clinical applications where labeled data is scarce. Our code is available at: https://github.com/qimingfan10/SMART.
Authors: Longmi Gao, Pan Gao
Abstract: Volume Electron Microscopy (VEM) is crucial for 3D tissue imaging but often produces anisotropic data with poor axial resolution, hindering visualization and downstream analysis. Existing methods for isotropic reconstruction often suffer from neglecting abundant axial information and employing simple downsampling to simulate anisotropic data. To address these limitations, we propose VEMamba, an efficient framework for isotropic reconstruction. The core of VEMamba is a novel 3D Dependency Reordering paradigm, implemented via two key components: an Axial-Lateral Chunking Selective Scan Module (ALCSSM), which intelligently re-maps complex 3D spatial dependencies (both axial and lateral) into optimized 1D sequences for efficient Mamba-based modeling, explicitly enforcing axial-lateral consistency; and a Dynamic Weights Aggregation Module (DWAM) to adaptively aggregate these reordered sequence outputs for enhanced representational power. Furthermore, we introduce a realistic degradation simulation and then leverage Momentum Contrast (MoCo) to integrate this degradation-aware knowledge into the network for superior reconstruction. Extensive experiments on both simulated and real-world anisotropic VEM datasets demonstrate that VEMamba achieves highly competitive performance across various metrics while maintaining a lower computational footprint. The source code is available on GitHub: https://github.com/I2-Multimedia-Lab/VEMamba
Authors: Zhanpeng Luo, Ce Zhang, Silong Yong, Cunxi Dai, Qianwei Wang, Haoxi Ran, Guanya Shi, Katia Sycara, Yaqi Xie
Abstract: Multi-modal Large Language Models (MLLMs) have demonstrated strong capabilities in general-purpose perception and reasoning, but they still struggle with tasks that require spatial understanding of the 3D world. To address this, we introduce pySpatial, a visual programming framework that equips MLLMs with the ability to interface with spatial tools via Python code generation. Given an image sequence and a natural-language query, the model composes function calls to spatial tools including 3D reconstruction, camera-pose recovery, novel-view rendering, etc. These operations convert raw 2D inputs into an explorable 3D scene, enabling MLLMs to reason explicitly over structured spatial representations. Notably, pySpatial requires no gradient-based fine-tuning and operates in a fully zero-shot setting. Experimental evaluations on the challenging MindCube and Omni3D-Bench benchmarks demonstrate that our framework pySpatial consistently surpasses strong MLLM baselines; for instance, it outperforms GPT-4.1-mini by 12.94% on MindCube. Furthermore, we conduct real-world indoor navigation experiments where the robot can successfully traverse complex environments using route plans generated by pySpatial, highlighting the practical effectiveness of our approach.
Authors: Xiaolong Zeng, Yitong Yu, Shiyao Xiong, Jinhua Hao, Ming Sun, Chao Zhou, Bin Wang
Abstract: Look-Up Table based methods have emerged as a promising direction for efficient image restoration tasks. Recent LUT-based methods focus on improving their performance by expanding the receptive field. However, they inevitably introduce extra computational and storage overhead, which hinders their deployment in edge devices. To address this issue, we propose ShiftLUT, a novel framework that attains the largest receptive field among all LUT-based methods while maintaining high efficiency. Our key insight lies in three complementary components. First, Learnable Spatial Shift module (LSS) is introduced to expand the receptive field by applying learnable, channel-wise spatial offsets on feature maps. Second, we propose an asymmetric dual-branch architecture that allocates more computation to the information-dense branch, substantially reducing inference latency without compromising restoration quality. Finally, we incorporate a feature-level LUT compression strategy called Error-bounded Adaptive Sampling (EAS) to minimize the storage overhead. Compared to the previous state-of-the-art method TinyLUT, ShiftLUT achieves a 3.8$\times$ larger receptive field and improves an average PSNR by over 0.21 dB across multiple standard benchmarks, while maintaining a small storage size and inference time.
Authors: Puyun Wang, Kaimin Yu, Huayang He, Feng Huang, Xianyu Wu, Yating Chen
Abstract: Underwater optical imaging is severely hindered by scattering, but polarization imaging offers the unique dual advantages of descattering and shape-from-polarization (SfP) 3D reconstruction. To exploit these advantages, this paper proposes UD-SfPNet, an underwater descattering shape-from-polarization network that leverages polarization cues for improved 3D surface normal prediction. The framework jointly models polarization-based image descattering and SfP normal estimation in a unified pipeline, avoiding error accumulation from sequential processing and enabling global optimization across both tasks. UD-SfPNet further incorporates a novel color embedding module to enhance geometric consistency by exploiting the relationship between color encodings and surface orientation. A detail enhancement convolution module is also included to better preserve high-frequency geometric details that are lost under scattering. Experiments on the MuS-Polar3D dataset show that the proposed method significantly improves reconstruction accuracy, achieving a mean surface normal angular error of 15.12$^\circ$ (the lowest among compared methods). These results confirm the efficacy of combining descattering with polarization-based shape inference, and highlight the practical significance and potential applications of UD-SfPNet for optical 3D imaging in challenging underwater environments. The code is available at https://github.com/WangPuyun/UD-SfPNet.
Authors: Sushish Baral, Paulo Garcia, Warisa Sritriratanarak
Abstract: The identification of repeating patterns in discrete grids is rudimentary within symbolic reasoning, algorithm synthesis and structural optimization across diverse computational domains. Although statistical approaches targeting noisy data can approximately recognize patterns, symbolic analysis utilizing deterministic extraction of periodic structures is underdeveloped. This paper aims to fill this gap by employing a hierarchical algorithm that discovers exact tessellations in finite planar grids, addressing the problem where multiple independent patterns may coexist within a hierarchical structure. The proposed method utilizes composite discovery (dual inspection and breadth-first pruning) for identifying rectangular regions with internal repetition, normalization to a minimal representative form, and prime extraction (selective duplication and hierarchical memoization) to account for irregular dimensions and to achieve efficient computation time. We evaluate scalability on grid sizes from 2x2 to 32x32, showing overlap detection on simple repeating tiles exhibits processing time under 1ms, while complex patterns which require exhaustive search and systematic exploration shows exponential growth. This algorithm provides deterministic behavior for exact, axis-aligned, rectangular tessellations, addressing a critical gap in symbolic grid analysis techniques, applicable to puzzle solving reasoning tasks and identification of exact repeating structures in discrete symbolic domains.
Authors: Yang Cao, Feize Wu, Dave Zhenyu Chen, Yingji Zhong, Lanqing Hong, Dan Xu
Abstract: Current multi-view indoor 3D object detectors rely on sensor geometry that is costly to obtain (i.e., precisely calibrated multi-view camera poses) to fuse multi-view information into a global scene representation, limiting deployment in real-world scenes. We target a more practical setting: Sensor-Geometry-Free (SG-Free) multi-view indoor 3D object detection, where there are no sensor-provided geometric inputs (multi-view poses or depth). Recent Visual Geometry Grounded Transformer (VGGT) shows that strong 3D cues can be inferred directly from images. Building on this insight, we present VGGT-Det, the first framework tailored for SG-Free multi-view indoor 3D object detection. Rather than merely consuming VGGT predictions, our method integrates VGGT encoder into a transformer-based pipeline. To effectively leverage both the semantic and geometric priors from inside VGGT, we introduce two novel key components: (i) Attention-Guided Query Generation (AG): exploits VGGT attention maps as semantic priors to initialize object queries, improving localization by focusing on object regions while preserving global spatial structure; (ii) Query-Driven Feature Aggregation (QD): a learnable See-Query interacts with object queries to 'see' what they need, and then dynamically aggregates multi-level geometric features across VGGT layers that progressively lift 2D features into 3D. Experiments show that VGGT-Det significantly surpasses the best-performing method in the SG-Free setting by 4.4 and 8.6 mAP@0.25 on ScanNet and ARKitScenes, respectively. Ablation study shows that VGGT's internally learned semantic and geometric priors can be effectively leveraged by our AG and QD.
Authors: Seungwook Kim, Minsu Cho
Abstract: Text-to-image generation powers content creation across design, media, and data augmentation. Post-training of text-to-image generative models is a promising path to better match human preferences, factuality, and improved aesthetics. We introduce ARC (Adaptive Rewarding by self-Confidence), a post-training framework that replaces external reward supervision with an internal self-confidence signal, obtained by evaluating how accurately the model recovers injected noise under self-denoising probes. ARC converts this intrinsic signal into scalar rewards, enabling fully unsupervised optimization without additional datasets, annotators, or reward models. Empirically, by reinforcing high-confidence generations, ARC delivers consistent gains in compositional generation, text rendering and text-image alignment over the baseline. We also find that integrating ARC with external rewards results in a complementary improvement, with alleviated reward hacking.
Authors: Zhiye Wang, Yanbo Jiang, Rui Zhou, Bo Zhang, Fang Zhang, Zhenhua Xu, Yaqin Zhang, Jianqiang Wang
Abstract: Large language models (LLMs) have shown great promise for autonomous driving. However, discretizing numbers into tokens limits precise numerical reasoning, fails to reflect the positional significance of digits in the training objective, and makes it difficult to achieve both decoding efficiency and numerical precision. These limitations affect both the processing of sensor measurements and the generation of precise control commands, creating a fundamental barrier for deploying LLM-based autonomous driving systems. In this paper, we introduce DriveCode, a novel numerical encoding method that represents numbers as dedicated embeddings rather than discrete text tokens. DriveCode employs a number projector to map numbers into the language model's hidden space, enabling seamless integration with visual and textual features in a unified multimodal sequence. Evaluated on OmniDrive, DriveGPT4, and DriveGPT4-V2 datasets, DriveCode demonstrates superior performance in trajectory prediction and control signal generation, confirming its effectiveness for LLM-based autonomous driving systems.
Authors: Md. Adnanul Islam, Wasimul Karim, Md Mahbub Alam, Subhey Sadi Rahman, Md. Abdur Rahman, Arefin Ittesafun Abian, Mohaimenul Azam Khan Raiaan, Kheng Cher Yeo, Deepika Mathur, Sami Azam
Abstract: Accurate weight estimation of commercial and industrial waste is important for efficient operations, yet image-based estimation remains difficult because similar-looking objects may have different densities, and the visible size changes with camera distance. Addressing this problem, we propose Multimodal Weight Predictor (MWP) framework that estimates waste weight by combining RGB images with physics-informed metadata, including object dimensions, camera distance, and camera height. We also introduce Waste-Weight-10K, a real-world dataset containing 10,421 synchronized image-metadata collected from logistics and recycling sites. The dataset covers 11 waste categories and a wide weight range from 3.5 to 3,450 kg. Our model uses a Vision Transformer for visual features and a dedicated metadata encoder for geometric and category information, combining them with Stacked Mutual Attention Fusion that allows visual and physical cues guide each other. This helps the model manage perspective effects and link objects to material properties. To ensure stable performance across the wide weight range, we train the model using Mean Squared Logarithmic Error. On the test set, the proposed method achieves 88.06 kg Mean Absolute Error (MAE), 6.39% Mean Absolute Percentage Error (MAPE), and an R2 coefficient of 0.9548. The model shows strong accuracy for light objects in the 0-100 kg range with 2.38 kg MAE and 3.1% MAPE, maintaining reliable performance for heavy waste in the 1000-2000 kg range with 11.1% MAPE. Finally, we incorporate a physically grounded explanation module using Shapley Additive Explanations (SHAP) and a large language model to provide clear, human-readable explanations for each prediction.
Authors: Shreshth Saini, Bowen Chen, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
Abstract: High Dynamic Range (HDR) user-generated (UGC) videos are rapidly proliferating across social platforms, yet most perceptual video quality assessment (VQA) systems remain tailored to Standard Dynamic Range (SDR). HDR has a higher bit depth, wide color gamut, and elevated luminance range, exposing distortions such as near-black crushing, highlight clipping, banding, and exposure flicker that amplify UGC artifacts and challenge SDR models. To catalyze progress, we curate Beyond8Bits, a large-scale subjective dataset of 44K videos from 6.5K sources with over 1.5M crowd ratings, spanning diverse scenes, capture conditions, and compression settings. We further introduce HDR-Q, the first Multimodal Large Language Model (MLLM) for HDR-UGC VQA. We propose (i) a novel HDR-aware vision encoder to produce HDR-sensitive embeddings, and (ii) HDR-Aware Policy Optimization (HAPO), an RL finetuning framework that anchors reasoning to HDR cues. HAPO augments GRPO via an HDR-SDR contrastive KL that encourages token reliance on HDR inputs and a Gaussian weighted regression reward for fine-grained MOS calibration. Across Beyond8Bits and public HDR-VQA benchmarks, HDR-Q delivers state-of-the-art performance.
Authors: Zhenchen Wan, Ce Chen, Runqi Lin, Jiaxin Huang, Tianxi Chen, Yanwu Xu, Tongliang Liu, Mingming Gong
Abstract: Virtual try-on (VTON) has recently achieved impressive visual fidelity, but most existing systems require uploading personal photos to cloud-based GPUs, raising privacy concerns and limiting on-device deployment. To address this, we present \textsc{Mobile-VTON}, a high-quality, privacy-preserving framework that enables fully offline virtual try-on on commodity mobile devices using only a single user image and a garment image. \textsc{Mobile-VTON} introduces a modular TeacherNet--GarmentNet--TryonNet (TGT) architecture that integrates knowledge distillation, garment-conditioned generation, and garment alignment into a unified pipeline optimized for on-device efficiency. Within this framework, we propose a Feature-Guided Adversarial (FGA) Distillation strategy that combines teacher supervision with adversarial learning to better match real-world image distributions. GarmentNet is trained with a trajectory-consistency loss to preserve garment semantics across diffusion steps, while TryonNet uses latent concatenation and lightweight cross-modal conditioning to enable robust garment-to-person alignment without large-scale pretraining. By combining these components, \textsc{Mobile-VTON} achieves high-fidelity generation with low computational overhead. Experiments on VITON-HD and DressCode at $1024{\times}768$ show that it matches or outperforms strong server-based baselines while running entirely offline. These results demonstrate that high-quality VTON is not only feasible but also practical on-device, offering a secure solution for real-world applications.
Authors: Wenxiang Jiang, Yujun Lan, Shuo Zhao, Yuanshan Liu, Mingzhu Zhou, Jinxin Wang
Abstract: Recently, Instant Neural Graphics Primitives (Instant-NGP) has achieved significant success in rapid 3D scene reconstruction, but securely embedding high-capacity hidden data, such as an entire 3D scene, remains a challenge. Existing methods rely on external decoders, require architectural modifications, and suffer from limited capacity, which makes them easily detectable. We propose a novel parameter-free 3D Cryptographic Steganography using Instant-NGP (StegoNGP), which leverages the Instant-NGP hash encoding function as a key-controlled scene switcher. By associating a default key with a cover scene and a secret key with a hidden scene, our method trains a single model to interweave both representations within the same network weights. The resulting model is indistinguishable from a standard Instant-NGP in architecture and parameter count. We also introduce an enhanced Multi-Key scheme, which assigns multiple independent keys across hash levels, dramatically expanding the key space and providing high robustness against partial key disclosure attacks. Experimental results demonstrated that StegoNGP can hide a complete high-quality 3D scene with strong imperceptibility and security, providing a new paradigm for high-capacity, undetectable information hiding in neural fields. The code can be found at https://github.com/jiang-wenxiang/StegoNGP.
Authors: Yi Zhang, Yulei Kang, Jian-Fang Hu
Abstract: High-fidelity reconstruction of dynamic scenes is an important yet challenging problem. While recent 4D Gaussian Splatting (4DGS) has demonstrated the ability to model temporal dynamics, it couples Gaussian motion and geometric attributes within a single covariance formulation, which limits its expressiveness for complex motions and often leads to visual artifacts. To address this, we propose VeGaS, a novel velocity-based 4D Gaussian Splatting framework that decouples Gaussian motion and geometry. Specifically, we introduce a Galilean shearing matrix that explicitly incorporates time-varying velocity to flexibly model complex non-linear motions, while strictly isolating the effects of Gaussian motion from the geometry-related conditional Gaussian covariance. Furthermore, a Geometric Deformation Network is introduced to refine Gaussian shapes and orientations using spatio-temporal context and velocity cues, enhancing temporal geometric modeling. Extensive experiments on public datasets demonstrate that VeGaS achieves state-of-the-art performance.
Authors: Jiangshan Wang, Kang Zhao, Jiayi Guo, Jiayu Wang, Hang Guo, Chenyang Zhu, Xiu Li, Xiangyu Yue
Abstract: High computational costs and slow inference hinder the practical application of video generation models. While prior works accelerate the generation process through feature caching, they often suffer from notable quality degradation. In this work, we reveal that this issue arises from their inability to distinguish truly redundant features, which leads to the unintended skipping of computations on important features. To address this, we propose \textbf{PreciseCache}, a plug-and-play framework that precisely detects and skips truly redundant computations, thereby accelerating inference without sacrificing quality. Specifically, PreciseCache contains two components: LFCache for step-wise caching and BlockCache for block-wise caching. For LFCache, we compute the Low-Frequency Difference (LFD) between the prediction features of the current step and those from the previous cached step. Empirically, we observe that LFD serves as an effective measure of step-wise redundancy, accurately detecting highly redundant steps whose computation can be skipped through reusing cached features. To further accelerate generation within each non-skipped step, we propose BlockCache, which precisely detects and skips redundant computations at the block level within the network. Extensive experiments on various backbones demonstrate the effectiveness of our PreciseCache, which achieves an average of 2.6x speedup without noticeable quality loss. Source code will be released.
Authors: Zhaoxin Fan, Nanxiang Jiang, Daiheng Gao, Shiji Zhou, Wenjun Wu
Abstract: Removing undesired concepts from large-scale text-to-image (T2I) and text-to-video (T2V) diffusion models while preserving overall generative quality remains a major challenge, particularly as modern models such as Stable Diffusion v3, Flux, and OpenSora employ flow-matching and transformer-based architectures and extend to long-horizon video generation. Existing concept erasure methods, designed for earlier T2I/T2V models, often fail to generalize to these paradigms. To address this issue, we propose EraseAnything++, a unified framework for concept erasure in both image and video diffusion models with flow-matching objectives. Central to our approach is formulating concept erasure as a constrained multi-objective optimization problem that explicitly balances concept removal with preservation of generative utility. To solve the resulting conflicting objectives, we introduce an efficient utility-preserving unlearning strategy based on implicit gradient surgery. Furthermore, by integrating LoRA-based parameter tuning with attention-level regularization, our method anchors erasure on key visual representations and propagates it consistently across spatial and temporal dimensions. In the video setting, we further enhance consistency through an anchor-and-propagate mechanism that initializes erasure on reference frames and enforces it throughout subsequent transformer layers, thereby mitigating temporal drift. Extensive experiments on both image and video benchmarks demonstrate that EraseAnything++ substantially outperforms prior methods in erasure effectiveness, generative fidelity, and temporal consistency, establishing a new state of the art for concept erasure in next-generation diffusion models.
Authors: Jiaqi Tang, Mengyan Zheng, Shu Zhang, Fandong Zhang, Qingchao Chen
Abstract: Vision Transformers (ViTs) excel in 3D medical segmentation but require massive annotated datasets. While Self-Supervised Learning (SSL) mitigates this using unlabeled data, it still faces strict privacy and logistical barriers. Formula-Driven Supervised Learning (FDSL) offers a privacy-preserving alternative by pre-training on synthetic mathematical primitives. However, a critical semantic gap limits its efficacy: generic shapes lack the morphological fidelity, fixed spatial layouts, and inter-organ relationships of real anatomy, preventing models from learning essential global structural priors. To bridge this gap, we propose an Anatomy-Informed Synthetic Supervised Pre-training framework unifying FDSL's infinite scalability with anatomical realism. We replace basic primitives with a lightweight shape bank with de-identified, label-only segmentation masks from 5 subjects. Furthermore, we introduce a structure-aware sequential placement strategy to govern the patch synthesis process. Instead of random placement, we enforce physiological plausibility using spatial anchors for correct localization and a topological graph to manage inter-organ interactions (e.g., preventing impossible overlaps). Extensive experiments on BTCV and MSD datasets demonstrate that our method significantly outperforms state-of-the-art FDSL baselines and SSL methods by 1.74\% and up to 1.66\%, while exhibiting a robust scaling effect where performance improves with increased synthetic data volume. This provides a data-efficient, privacy-compliant solution for medical segmentation. The code will be made publicly available upon acceptance.
Authors: Wang Chen, Yongdong Luo, Yuhui Zeng, Luojun Lin, Tianyu Xie, Fei Chao, Rongrong Ji, Xiawu Zheng
Abstract: Massive frame redundancy and limited context window make efficient frame selection crucial for long-video understanding with large vision-language models (LVLMs). Prevailing approaches, however, adopt a flat sampling paradigm which treats the video as an unstructured collection of frames. In this paper, we introduce Event-Anchored Frame Selection (EFS), a hierarchical, event-aware pipeline. Leveraging self-supervised DINO embeddings, EFS first partitions the video stream into visually homogeneous temporal segments, which serve as proxies for semantic events. Within each event, it then selects the most query-relevant frame as an anchor. These anchors act as structural priors that guide a global refinement stage using an adaptive Maximal Marginal Relevance (MMR) scheme. This pipeline ensures the final keyframe set jointly optimizes for event coverage, query relevance, and visual diversity. As a training-free, plug-and-play module, EFS can be seamlessly integrated into off-the-shelf LVLMs, yielding substantial gains on challenging video understanding benchmarks. Specifically, when applied to LLaVA-Video-7B, EFS improves accuracy by 4.7%, 4.9%, and 8.8% on VideoMME, LongVideoBench, and MLVU, respectively.
Authors: Jiaqi Tang, Weixuan Xu, Shu Zhang, Fandong Zhang, Qingchao Chen
Abstract: Vision Transformers (ViTs) have revolutionized medical image analysis, yet their data-hungry nature clashes with the scarcity and privacy constraints of clinical archives. Formula-Driven Supervised Learning (FDSL) has emerged as a promising solution to this bottleneck, synthesizing infinite annotated samples from mathematical formulas without utilizing real patient data. However, existing FDSL paradigms rely on simple geometric shapes with homogeneous intensities, creating a substantial gap by neglecting tissue textures and noise patterns inherent in modalities like CT and MRI. In this paper, we identify a critical optimization conflict termed boundary aliasing: when high-frequency synthetic textures are naively added, they corrupt the image gradient signals necessary for learning structural boundaries, causing the model to fail in delineating real anatomical margins. To bridge this gap, we propose a novel Physics-inspired Spatially-Decoupled Synthesis framework. Our approach orthogonalizes the synthesis process: it first constructs a gradient-shielded buffer zone based on boundary distance to ensure stable shape learning, and subsequently injects physics-driven spectral textures into the object core. This design effectively reconciles robust shape representation learning with invariance to acquisition noise. Extensive experiments on the BTCV and MSD datasets demonstrate that our method significantly outperforms previous FDSL, as well as SSL methods trained on real-world medical datasets, by 1.43% on BTCV and up to 1.51% on MSD task, offering a scalable, annotation-free foundation for medical ViTs. The code will be made publicly available upon acceptance.
Authors: Danfeng Hong, Chenyu Li, Xuyang Li, Gustau Camps-Valls, Jocelyn Chanussot
Abstract: Remote sensing (RS) techniques are increasingly crucial for deepening our understanding of the planet. As the volume and diversity of RS data continue to grow exponentially, there is an urgent need for advanced data modeling and understanding capabilities to manage and interpret these vast datasets effectively. Foundation models present significant new growth opportunities and immense potential to revolutionize the RS field. In this paper, we conduct a comprehensive technical survey on foundation models in RS, offering a brand-new perspective by exploring their evolution from unimodality to multimodality. We hope this work serves as a valuable entry point for researchers interested in both foundation models and RS and helps them launch new projects or explore new research topics in this rapidly evolving area. This survey addresses the following three key questions: What are foundation models in RS? Why are foundation models needed in RS? How can we effectively guide junior researchers in gaining a comprehensive and practical understanding of foundation models in RS applications? More specifically, we begin by outlining the background and motivation, emphasizing the importance of foundation models in RS. We then review existing foundation models in RS, systematically categorizing them into unimodal and multimodal approaches. Additionally, we provide a tutorial-like section to guide researchers, especially beginners, on how to train foundation models in RS and apply them to real-world tasks. The survey aims to equip researchers in RS with a deeper and more efficient understanding of foundation models, enabling them to get started easily and effectively apply these models across various RS applications.
Authors: Yi Zhang, Puxun Tu, Kun Wang, Yulin Yan, Tao Ying, Xiaojun Chen
Abstract: Freehand 3D ultrasound (US) reconstruction promises volumetric imaging with the flexibility of standard 2D probes, yet existing tracking paradigms face a restrictive trilemma: marker-based systems demand prohibitive costs, inside-out methods require intrusive sensor attachment, and sensorless approaches suffer from severe cumulative drift. To overcome these limitations, we present MLRecon, a robust markerless 3D US reconstruction framework delivering drift-resilient 6D probe pose tracking using a single commodity RGB-D camera. Leveraging the generalization power of vision foundation models, our pipeline enables continuous markerless tracking of the probe, augmented by a vision-guided divergence detector that autonomously monitors tracking integrity and triggers failure recovery to ensure uninterrupted scanning. Crucially, we further propose a dual-stage pose refinement network that explicitly disentangles high-frequency jitter from low-frequency bias, effectively denoising the trajectory while maintaining the kinematic fidelity of operator maneuvers. Experiments demonstrate that MLRecon significantly outperforms competing sensorless and sensor-aided methods, achieving average position errors as low as 0.88 mm on complex trajectories and yielding high-quality 3D reconstructions with sub-millimeter mean surface accuracy. This establishes a new benchmark for low-cost, accessible volumetric US imaging in resource-limited clinical settings.
Authors: Yuze Li, Dong Gong, Xiao Cao, Junchao Yuan, Dongsheng Li, Lei Zhou, Yun Sing Koh, Cheng Yan, Xinyu Zhang
Abstract: Motion transfer has emerged as a promising direction for controllable video generation, yet existing methods largely focus on single-object scenarios and struggle when multiple objects require distinct motion patterns. In this work, we present FlexiMMT, the first implicit image-to-video (I2V) motion transfer framework that explicitly enables multi-object, multi-motion transfer. Given a static multi-object image and multiple reference videos, FlexiMMT independently extracts motion representations and accurately assigns them to different objects, supporting flexible recombination and arbitrary motion-to-object mappings. To address the core challenge of cross-object motion entanglement, we introduce a Motion Decoupled Mask Attention Mechanism that uses object-specific masks to constrain attention, ensuring that motion and text tokens only influence their designated regions. We further propose a Differentiated Mask Propagation Mechanism that derives object-specific masks directly from diffusion attention and progressively propagates them across frames efficiently. Extensive experiments demonstrate that FlexiMMT achieves precise, compositional, and state-of-the-art performance in I2V-based multi-object multi-motion transfer.
Authors: Xubo Zhu, Haoyang Zhang, Fei He, Rui Wu, Yanhu Shan, Wen Yang, Huai Yu
Abstract: 3D semantic occupancy prediction is crucial for autonomous driving perception, offering comprehensive geometric scene understanding and semantic recognition. However, existing methods struggle with geometric misalignment in view transformation due to the lack of pixel-level accurate depth estimation, and severe spatial class imbalance where semantic categories exhibit strong spatial anisotropy. To address these challenges, we propose Dr.Occ, a depth- and region-guided occupancy prediction framework. Specifically, we introduce a depth-guided 2D-to-3D View Transformer (D$^2$-VFormer) that effectively leverages high-quality dense depth cues from MoGe-2 to construct reliable geometric priors, thereby enabling precise geometric alignment of voxel features. Moreover, inspired by the Mixture-of-Experts (MoE) framework, we propose a region-guided Expert Transformer (R/R$^2$-EFormer) that adaptively allocates region-specific experts to focus on different spatial regions, effectively addressing spatial semantic variations. Thus, the two components make complementary contributions: depth guidance ensures geometric alignment, while region experts enhance semantic learning. Experiments on the Occ3D-nuScenes benchmark demonstrate that \textbf{Dr.Occ} improves the strong baseline BEVDet4D by 7.43\% mIoU and 3.09\% IoU under the full vision-only setting.
Authors: Xuqin Wang, Tao Wu, Yanfeng Zhang, Lu Liu, Mingwei Sun, Yongliang Wang, Niclas Zeller, Daniel Cremers
Abstract: Recent advances in generative modeling have substantially enhanced novel view synthesis, yet maintaining consistency across viewpoints remains challenging. Diffusion-based models rely on stochastic noise-to-data transitions, which obscure deterministic structures and yield inconsistent view predictions. We propose a Data-to-Data Flow Matching framework that learns deterministic transformations directly between paired views, enhancing view-consistent synthesis through explicit data coupling. To further enhance geometric coherence, we introduce Probability Density Geodesic Flow Matching (PDG-FM), which constrains flow trajectories using geodesic interpolants derived from probability density metrics of pretrained diffusion models. Such alignment with high-density regions of the data manifold promotes more realistic interpolants between samples. Empirically, our method surpasses diffusion-based NVS baselines, demonstrating improved structural coherence and smoother transitions across views. These results highlight the advantages of incorporating data-dependent geometric regularization into deterministic flow matching for consistent novel view generation.
Authors: Yiquan Gao
Abstract: Car license plate recognition system is an image processing technology used to identify vehicles by capturing their Car License Plates. The car license plate recognition technology is also known as automatic number-plate recognition, automatic vehicle identification, car license plate recognition or optical character recognition for cars. In Malaysia, as the number of vehicle is increasing rapidly nowadays, a pretty great number of vehicle on the road has brought about the considerable demands of car license plate recognition system. Car license plate recognition system can be implemented in electronic parking payment system, highway toll-fee system, traffic surveillance system and as police enforcement tools. Additionally, car license plate recognition system technology also has potential to be combined with various techniques in other different fields like biology, aerospace and so on to achieve the goal of solving some specialized problems.
Authors: Shengpeng Wang, Kuangyu Wang, Wei Wang
Abstract: Millimeter-wave radar offers unique advantages in adverse weather but suffers from low spatial fidelity, severe azimuth ambiguity, and clutter-induced spurious returns. Existing methods mainly focus on improving spatial perception effectiveness via coarse-to-fine cross-modal supervision, yet often overlook the ambiguous feature-to-label mapping, which may lead to ill-posed geometric inference and pose fundamental challenges to downstream perception tasks. In this work, we propose RaUF, a spatial uncertainty field learning framework that models radar measurements through their physically grounded anisotropic properties. To resolve conflicting feature-to-label mapping, we design an anisotropic probabilistic model that learns fine-grained uncertainty. To further enhance reliability, we propose a Bidirectional Domain Attention mechanism that exploits the mutual complementarity between spatial structure and Doppler consistency, effectively suppressing spurious or multipath-induced reflections. Extensive experiments on public benchmarks and real-world datasets demonstrate that RaUF delivers highly reliable spatial detections with well-calibrated uncertainty. Moreover, downstream case studies further validate the enhanced reliability and scalability of RaUF under challenging real-world driving scenarios.
Authors: Junbo Ke, Yangyang Xu, You-Wei Wen, Chao Wang
Abstract: Implicit Neural Representations (INRs) have emerged as a powerful paradigm for various signal processing tasks, but their inherent spectral bias limits the ability to capture high-frequency details. Existing methods partially mitigate this issue by using Fourier-based features, which usually rely on fixed frequency bases. This forces multi-layer perceptrons (MLPs) to inefficiently compose the required frequencies, thereby constraining their representational capacity. To address this limitation, we propose Content-Aware Frequency Encoding (CAFE), which builds upon Fourier features through multiple parallel linear layers combined via a Hadamard product. CAFE can explicitly and efficiently synthesize a broader range of frequency bases, while the learned weights enable the selection of task-relevant frequencies. Furthermore, we extend this framework to CAFE+, which incorporates Chebyshev features as a complementary component to Fourier bases. This combination provides a stronger and more stable frequency representation. Extensive experiments across multiple benchmarks validate the effectiveness and efficiency of our approach, consistently achieving superior performance over existing methods. Our code is available at https://github.com/JunboKe0619/CAFE.
Authors: Zhuolin He, Jiacheng Tang, Jian Pu, Xiangyang Xue
Abstract: Safe autonomous systems in complex environments require robust road anomaly segmentation to identify unknown obstacles. However, existing approaches often rely on pixel-level statistics to determine whether a region appears anomalous. This reliance leads to high false-positive rates on semantically normal background regions such as sky or vegetation, and poor recall of true Out-of-distribution (OOD) instances, thereby posing safety risks for robotic perception and decision-making. To address these challenges, we propose VL-Anomaly, a vision-language anomaly segmentation framework that incorporates semantic priors from pre-trained Vision-Language Models (VLMs). Specifically, we design a prompt learning-driven alignment module that adapts Mask2Forme's visual features to CLIP text embeddings of known categories, effectively suppressing spurious anomaly responses in background regions. At inference time, we further introduce a multi-source inference strategy that integrates text-guided similarity, CLIP-based image-text similarity and detector confidence, enabling more reliable anomaly prediction by leveraging complementary information sources. Extensive experiments demonstrate that VL-Anomaly achieves state-of-the-art performance on benchmark datasets including RoadAnomaly, SMIYC and Fishyscapes.Code is released on https://github.com/NickHezhuolin/VL-aligner-Road-anomaly-segment.
URLs: https://github.com/NickHezhuolin/VL-aligner-Road-anomaly-segment.
Authors: Yangyang Xu, Junbo Ke, You-Wei Wen, Chao Wang
Abstract: Tensor Ring (TR) decomposition is a powerful tool for high-order data modeling, but is inherently restricted to discrete forms defined on fixed meshgrids. In this work, we propose a TR functional decomposition for both meshgrid and non-meshgrid data, where factors are parameterized by Implicit Neural Representations (INRs). However, optimizing this continuous framework to capture fine-scale details is intrinsically difficult. Through a frequency-domain analysis, we demonstrate that the spectral structure of TR factors determines the frequency composition of the reconstructed tensor and limits the high-frequency modeling capacity. To mitigate this, we propose a reparameterized TR functional decomposition, in which each TR factor is a structured combination of a learnable latent tensor and a fixed basis. This reparameterization is theoretically shown to improve the training dynamics of TR factor learning. We further derive a principled initialization scheme for the fixed basis and prove the Lipschitz continuity of our proposed model. Extensive experiments on image inpainting, denoising, super-resolution, and point cloud recovery demonstrate that our method achieves consistently superior performance over existing approaches. Code is available at https://github.com/YangyangXu2002/RepTRFD.
Authors: Kuanxu Hou
Abstract: In robot automated assembly, snap assembly precision and efficiency directly determine overall production quality. As a core prerequisite, snap detection and localization critically affect subsequent assembly success. Traditional visual methods suffer from poor robustness and large localization errors when handling complex scenarios (e.g., transparent or low-contrast snaps), failing to meet high-precision assembly demands. To address this, this paper designs a dedicated sensor and proposes SMR-Net, an self-attention-based multi-scale object detection algorithm, to synergistically enhance detection and localization performance. SMR-Net adopts an attention-enhanced multi-scale feature fusion architecture: raw sensor data is encoded via an attention-embedded feature extractor to strengthen key snap features and suppress noise; three multi-scale feature maps are processed in parallel with standard and dilated convolution for dimension unification while preserving resolution; an adaptive reweighting network dynamically assigns weights to fused features, generating fine representations integrating details and global semantics. Experimental results on Type A and Type B snap datasets show SMR-Net outperforms traditional Faster R-CNN significantly: Intersection over Union (IoU) improves by 6.52% and 5.8%, and mean Average Precision (mAP) increases by 2.8% and 1.5% respectively. This fully demonstrates the method's superiority in complex snap detection and localization tasks.
Authors: Haoyuan Zhang, Keyao Wang, Guosheng Zhang, Haixiao Yue, Zhiwen Tan, Siran Peng, Tianshuo Zhang, Xiao Tan, Kunbin Chen, Wei He, Jingdong Wang, Ajian Liu, Xiangyu Zhu, Zhen Lei
Abstract: Face recognition remains vulnerable to presentation attacks, calling for robust Face Anti-Spoofing (FAS) solutions. Recent MLLM-based FAS methods reformulate the binary classification task as the generation of brief textual descriptions to improve cross-domain generalization. However, their generalizability is still limited, as such descriptions mainly capture intuitive semantic cues (e.g., mask contours) while struggling to perceive fine-grained visual patterns. To address this limitation, we incorporate external visual tools into MLLMs to encourage deeper investigation of subtle spoof clues. Specifically, we propose the Tool-Augmented Reasoning FAS (TAR-FAS) framework, which reformulates the FAS task as a Chain-of-Thought with Visual Tools (CoT-VT) paradigm, allowing MLLMs to begin with intuitive observations and adaptively invoke external visual tools for fine-grained investigation. To this end, we design a tool-augmented data annotation pipeline and construct the ToolFAS-16K dataset, which contains multi-turn tool-use reasoning trajectories. Furthermore, we introduce a tool-aware FAS training pipeline, where Diverse-Tool Group Relative Policy Optimization (DT-GRPO) enables the model to autonomously learn efficient tool use. Extensive experiments under a challenging one-to-eleven cross-domain protocol demonstrate that TAR-FAS achieves SOTA performance while providing fine-grained visual investigation for trustworthy spoof detection.
Authors: Huanjin Yao, Qixiang Yin, Min Yang, Ziwang Zhao, Yibo Wang, Haotian Luo, Jingyi Zhang, Jiaxing Huang
Abstract: We aim to develop a multimodal research agent capable of explicit reasoning and planning, multi-tool invocation, and cross-modal information synthesis, enabling it to conduct deep research tasks. However, we observe three main challenges in developing such agents: (1) scarcity of search-intensive multimodal QA data, (2) lack of effective search trajectories, and (3) prohibitive cost of training with online search APIs. To tackle them, we first propose Hyper-Search, a hypergraph-based QA generation method that models and connects visual and textual nodes within and across modalities, enabling to generate search-intensive multimodal QA pairs that require invoking various search tools to solve. Second, we introduce DR-TTS, which first decomposes search-involved tasks into several categories according to search tool types, and respectively optimize specialized search tool experts for each tool. It then recomposes tool experts to jointly explore search trajectories via tree search, producing trajectories that successfully solve complex tasks using various search tools. Third, we build an offline search engine supporting multiple search tools, enabling agentic reinforcement learning without using costly online search APIs. With the three designs, we develop MM-DeepResearch, a powerful multimodal deep research agent, and extensive results shows its superiority across benchmarks. Code is available at https://github.com/HJYao00/MM-DeepResearch
Authors: Yuechen Luo, Qimao Chen, Fang Li, Shaoqing Xu, Jaxin Liu, Ziying Song, Zhi-xin Yang, Fuxi Wen
Abstract: Vision-Language-Action (VLA) models for autonomous driving often hit a performance plateau during Reinforcement Learning (RL) optimization. This stagnation arises from exploration capabilities constrained by previous Supervised Fine-Tuning (SFT), leading to persistent failures in long-tail scenarios. In these critical situations, all explored actions yield a zero-value driving score. This information-sparse reward signals a failure, yet fails to identify its root cause -- whether it is due to incorrect planning, flawed reasoning, or poor trajectory execution. To address this limitation, we propose VLA with Explicit Learning from Failures (ELF-VLA), a framework that augments RL with structured diagnostic feedback. Instead of relying on a vague scalar reward, our method produces detailed, interpretable reports that identify the specific failure mode. The VLA policy then leverages this explicit feedback to generate a Feedback-Guided Refinement. By injecting these corrected, high-reward samples back into the RL training batch, our approach provides a targeted gradient, which enables the policy to solve critical scenarios that unguided exploration cannot. Extensive experiments demonstrate that our method unlocks the latent capabilities of VLA models, achieving state-of-the-art (SOTA) performance on the public NAVSIM benchmark for overall PDMS, EPDMS score and high-level planning accuracy.
Authors: Zebin You, Xiaolu Zhang, Jun Zhou, Chongxuan Li, Ji-Rong Wen
Abstract: We present \textbf{LLaDA-o}, an effective and length-adaptive omni diffusion model for multimodal understanding and generation. LLaDA-o is built on a Mixture of Diffusion (MoD) framework that decouples discrete masked diffusion for text understanding and continuous diffusion for visual generation, while coupling them through a shared, simple, and efficient attention backbone that reduces redundant computation for fixed conditions. Building on MoD, we further introduce a data-centric length adaptation strategy that enables flexible-length decoding in multimodal settings without architectural changes. Extensive experiments show that LLaDA-o achieves state-of-the-art performance among omni-diffusion models on multimodal understanding and generation benchmarks, and reaches 87.04 on DPG-Bench for text-to-image generation, supporting the effectiveness of unified omni diffusion modeling. Code is available at https://github.com/ML-GSAI/LLaDA-o.
Authors: Yunguan Fu, Wenjia Bai, Wen Yan, Matthew J Clarkson, Rhodri Huw Davies, Yipeng Hu
Abstract: Diffusion-based unsupervised image registration has been explored for cardiac cine MR, but expensive multi-step inference limits practical use. We propose FlowReg, a flow-matching framework in displacement field space that achieves strong registration in as few as two steps and supports further refinement with more steps. FlowReg uses warmup-reflow training: a single-step network first acts as a teacher, then a student learns to refine from arbitrary intermediate states, removing the need for a pre-trained model as in existing methods. An Initial Guess strategy feeds back the model prediction as the next starting point, improving refinement from step two onward. On ACDC and MM2 across six tasks (including cross-dataset generalization), FlowReg outperforms the state of the art on five tasks (+0.6% mean Dice score on average), with the largest gain in the left ventricle (+1.09%), and reduces LVEF estimation error on all six tasks (-2.58 percentage points), using only 0.7% extra parameters and no segmentation labels. Anonymized code is available at https://github.com/mathpluscode/FlowReg.
Authors: Wangkai Li, Zhaoyang Li, Yuwen Pan, Rui Sun, Yujia Chen, Tianzhu Zhang
Abstract: Adverse weather conditions significantly degrade the performance of LiDAR point cloud semantic segmentation networks by introducing large distribution shifts. Existing augmentation-based methods attempt to enhance robustness by simulating weather interference during training. However, they struggle to fully exploit the potential of augmentations due to the trade-off between minor and aggressive augmentations. To address this, we propose A3Point, an adaptive augmentation-aware latent learning framework that effectively utilizes a diverse range of augmentations while mitigating the semantic shift, which refers to the change in the semantic meaning caused by augmentations. A3Point consists of two key components: semantic confusion prior (SCP) latent learning, which captures the model's inherent semantic confusion information, and semantic shift region (SSR) localization, which decouples semantic confusion and semantic shift, enabling adaptive optimization strategies for different disturbance levels. Extensive experiments on multiple standard generalized LiDAR segmentation benchmarks under adverse weather demonstrate the effectiveness of our method, setting new state-of-the-art results.
Authors: Xuan Lu, Kangle Li, Haohang Huang, Rui Meng, Wenjun Zeng, Xiaoyu Shen
Abstract: Recent advances in multimodal large language models (MLLMs) have substantially expanded the capabilities of multimodal retrieval, enabling systems to align and retrieve information across visual and textual modalities. Yet, existing benchmarks largely focus on coarse-grained or single-condition alignment, overlooking real-world scenarios where user queries specify multiple interdependent constraints across modalities. To bridge this gap, we introduce MCMR (Multi-Conditional Multimodal Retrieval): a large-scale benchmark designed to evaluate fine-grained, multi-condition cross-modal retrieval under natural-language queries. MCMR spans five product domains: upper and bottom clothing, jewelry, shoes, and furniture. It also preserves rich long-form metadata essential for compositional matching. Each query integrates complementary visual and textual attributes, requiring models to jointly satisfy all specified conditions for relevance. We benchmark a diverse suite of MLLM-based multimodal retrievers and vision-language rerankers to assess their condition-aware reasoning abilities. Experimental results reveal: (i) distinct modality asymmetries across models; (ii) visual cues dominate early-rank precision, while textual metadata stabilizes long-tail ordering; and (iii) MLLM-based pointwise rerankers markedly improve fine-grained matching by explicitly verifying query-candidate consistency. Overall, MCMR establishes a challenging and diagnostic benchmark for advancing multimodal retrieval toward compositional, constraint-aware, and interpretable understanding. Our code and dataset is available at https://github.com/EIT-NLP/MCMR
Authors: Arctanx An, Shizhao Sun, Danqing Huang, Mingxi Cheng, Yan Gao, Ji Li, Yu Qiao, Jiang Bian
Abstract: Assessing the aesthetic quality of graphic design is central to visual communication, yet remains underexplored in vision language models (VLMs). We investigate whether VLMs can evaluate design aesthetics in ways comparable to humans. Prior work faces three key limitations: benchmarks restricted to narrow principles and coarse evaluation protocols, a lack of systematic VLM comparisons, and limited training data for model improvement. In this work, we introduce AesEval-Bench, a comprehensive benchmark spanning four dimensions, twelve indicators, and three fully quantifiable tasks: aesthetic judgment, region selection, and precise localization. Then, we systematically evaluate proprietary, open-source, and reasoning-augmented VLMs, revealing clear performance gaps against the nuanced demands of aesthetic assessment. Moreover, we construct a training dataset to fine-tune VLMs for this domain, leveraging human-guided VLM labeling to produce task labels at scale and indicator-grounded reasoning to tie abstract indicators to concrete design regions.Together, our work establishes the first systematic framework for aesthetic quality assessment in graphic design. Our code and dataset will be released at: \href{https://github.com/arctanxarc/AesEval-Bench}{https://github.com/arctanxarc/AesEval-Bench}
URLs: https://github.com/arctanxarc/AesEval-Bench, https://github.com/arctanxarc/AesEval-Bench
Authors: Yifu Qiu, Paul-Ambroise Duquenne, Holger Schwenk
Abstract: We introduce V-SONAR, a vision-language embedding space extended from the text-only embedding space SONAR (Omnilingual Embeddings Team et al., 2026), which supports 1500 text languages and 177 speech languages. To construct V-SONAR, we propose a post-hoc alignment pipeline that maps the representations of an existing vision encoder into the SONAR space. We thoroughly evaluate V-SONAR and show that its embeddings achieve competitive performance on text-to-video retrieval. Equipped with the OMNISONAR text decoder, V-SONAR further surpasses state-of-the-art vision-language models on video captioning tasks, including DREAM-1K (BLEU 23.9 vs. 19.6) and PE-VIDEO (BLEU 39.0 vs. 30.0). Leveraging V-SONAR, we first demonstrate that the Large Concept Model (LCM; LCM team et al. 2024) operating in SONAR and trained with English text only, can perform both single- and multi-visual concept understanding in a zero-shot manner. Finally, we introduce V-LCM, which extends the LCM with vision-language instruction tuning. V-LCM encodes vision and language inputs into an unified sequence of latent embeddings via V-SONAR and SONAR, and it is trained with the same latent diffusion objective for next-embedding prediction as in LCM's text-only pre-training. Experiments on a large-scale multilingual and -modal instruction-tuning data mixture highlight the potential of V-LCM: V-LCM matches state-of-the-art vision-language models on tasks covering image/video captioning and question answering, while significantly outperforming them across 61 rich- to low-resource languages out of all 62 tested languages.
Authors: Soroosh Tayebi Arasteh, Marziyeh Mohammadi, Sven Nebelung, Daniel Truhn
Abstract: Differential privacy (DP)'s effect in medical imaging is typically evaluated only through end-to-end performance, leaving the mechanism of privacy-induced utility loss unclear. We introduce Differential Privacy Representation Geometry for Medical Imaging (DP-RGMI), a framework that interprets DP as a structured transformation of representation space and decomposes performance degradation into encoder geometry and task-head utilization. Geometry is quantified by representation displacement from initialization and spectral effective dimension, while utilization is measured as the gap between linear-probe and end-to-end utility. Across over 594,000 images from four chest X-ray datasets and multiple pretrained initializations, we show that DP is consistently associated with a utilization gap even when linear separability is largely preserved. At the same time, displacement and spectral dimension exhibit non-monotonic, initialization- and dataset-dependent reshaping, indicating that DP alters representation anisotropy rather than uniformly collapsing features. Correlation analysis reveals that the association between end-to-end performance and utilization is robust across datasets but can vary by initialization, while geometric quantities capture additional prior- and dataset-conditioned variation. These findings position DP-RGMI as a reproducible framework for diagnosing privacy-induced failure modes and informing privacy model selection.
Authors: Jiashu Li, Xumeng Han, Zhaoyang Wei, Zipeng Wang, Kuiran Wang, Guorong Li, Zhenjun Han, Jianbin Jiao
Abstract: 3D Gaussian Splatting (3DGS) has recently emerged as a promising approach in novel view synthesis, combining photorealistic rendering with real-time efficiency. However, its success heavily relies on dense camera coverage; under sparse-view conditions, insufficient supervision leads to irregular Gaussian distributions, characterized by globally sparse coverage, blurred background, and distorted high-frequency areas. To address this, we propose HeroGS, Hierarchical Guidance for Robust 3D Gaussian Splatting, a unified framework that establishes hierarchical guidance across the image, feature, and parameter levels. At the image level, sparse supervision is converted into pseudo-dense guidance, globally regularizing the Gaussian distributions and forming a consistent foundation for subsequent optimization. Building upon this, Feature-Adaptive Densification and Pruning (FADP) at the feature level leverages low-level features to refine high-frequency details and adaptively densifies Gaussians in background regions. The optimized distributions then support Co-Pruned Geometry Consistency (CPG) at parameter level, which guides geometric consistency through parameter freezing and co-pruning, effectively removing inconsistent splats. The hierarchical guidance strategy effectively constrains and optimizes the overall Gaussian distributions, thereby enhancing both structural fidelity and rendering quality. Extensive experiments demonstrate that HeroGS achieves high-fidelity reconstructions and consistently surpasses state-of-the-art baselines under sparse-view conditions.
Authors: Dantong Qin, Alessandro Bozzon, Xian Yang, Xun Zhang, Yike Guo, Pan Wang
Abstract: Many creative multimedia systems are built upon visual primitives such as strokes or textures, which are difficult to collect at scale and fundamentally different from natural image data. This data scarcity makes it challenging for modern generative models to learn expressive and controllable primitives, limiting their use in process-aware content creation. In this work, we study the problem of learning human-like brushstroke generation from a small set of hand-drawn samples (n=470) and propose StrokeDiff, a diffusion-based framework with Smooth Regularization (SmR). SmR injects stochastic visual priors during training, providing a simple mechanism to stabilize diffusion models under sparse supervision without altering the inference process. We further show how the learned primitives can be made controllable through a B\'ezier-based conditioning module and integrated into a complete stroke-based painting pipeline, including prediction, generation, ordering, and compositing. This demonstrates how data-efficient primitive modeling can support expressive and structured multimedia content creation. Experiments indicate that the proposed approach produces diverse and structurally coherent brushstrokes and enables paintings with richer texture and layering, validated by both automatic metrics and human evaluation.
Authors: Tajamul Ashraf, Abrar Ul Riyaz, Wasif Tak, Tavaheed Tariq, Sonia Yadav, Moloud Abdar, Janibul Bashir
Abstract: Clinically reliable perception of surgical scenes is essential for advancing intelligent, context-aware intraoperative assistance such as instrument handoff guidance, collision avoidance, and workflow-aware robotic support. Existing surgical tool benchmarks primarily evaluate category-level segmentation, requiring models to detect all instances of predefined instrument classes. However, real-world clinical decisions often require resolving references to a specific instrument instance based on its functional role, spatial relation, or anatomical interaction capabilities not captured by current evaluation paradigms. We introduce GroundedSurg, the first language-conditioned, instance-level surgical grounding benchmark. Each instance pairs a surgical image with a natural-language description targeting a single instrument, accompanied by structured spatial grounding annotations including bounding boxes and point-level anchors. The dataset spans ophthalmic, laparoscopic, robotic, and open procedures, encompassing diverse instrument types, imaging conditions, and operative complexities. By jointly evaluating linguistic reference resolution and pixel-level localization, GroundedSurg enables a systematic and realistic evaluation of vision-language models in clinically realistic multi-instrument scenes. Extensive experiments demonstrate substantial performance gaps across modern segmentation and VLMs, highlighting the urgent need for clinically grounded vision-language reasoning in surgical AI systems. Code and data are publicly available at https://github.com/gaash-lab/GroundedSurg
Authors: Yiming Ma, Hongkun Yang, Lionel Z. Wang, Bin Chen, Weizhi Xian, Jianzhi Teng
Abstract: Prompt learning is a dominant paradigm for adapting pre-trained Vision-Language Models (VLMs) to downstream tasks. However, existing methods often rely on a simplistic, layer-centric view, assuming shallow layers capture general features while deep layers handle task-specific knowledge. This assumption results in uncontrolled interactions between learnable tokens and original tokens. Task-specific knowledge could degrades the model's core generalization and creates a trade-off between task adaptation and the preservation of zero-shot generalization. To address this, we challenge the layer-centric view and propose \textbf{DeAR}, a framework that achieves fine-grained VLM adaptation by \textbf{De}composing \textbf{A}ttention head \textbf{R}oles. We posit that the functional specialization within VLMs occurs not between layers, but at the finer-grained level of individual attention heads in the deeper layers. Based on this insight, we introduce a novel metric, Concept Entropy, to systematically classify attention heads into distinct functional roles: \textit{Attribute}, \textit{Generalization}, and \textit{Mixed}. Guided by these roles, we introduce specialized attribute tokens and a Role-Based Attention Mask mechanism to precisely control information flow, ensuring generalization heads remain isolated from task-specific knowledge. We further incorporate a Task-Adaptive Fusion Strategy for inference. Extensive experiments on fifteen datasets show that DeAR achieves a strong balance between task adaptation and generalization, outperforming previous methods across various tasks.
Authors: Zhuonan Liang, Wei Guo, Jie Gan, Yaxuan Song, Runnan Chen, Hang Chang, Weidong Cai
Abstract: Foundation vision models are increasingly adopted in medical image analysis. Due to domain shift, these pretrained models misalign with medical image segmentation needs without being fully fine-tuned or lightly adapted. We introduce GuiDINO, a framework that repositions native foundation model to acting as a visual guidance generator for downstream segmentation. GuiDINO extracts visual feature representation from DINOv3 and converts them into a spatial guide mask via a lightweight TokenBook mechanism, which aggregates token-prototype similarities. This guide mask gates feature activations in multiple segmentation backbones, thereby injecting foundation-model priors while preserving the inductive biases and efficiency of medical dedicated architectures. Training relies on a guide supervision objective loss that aligns the guide mask to ground-truth regions, optionally augmented by a boundary-focused hinge loss to sharpen fine structures. GuiDINO also supports parameter-efficient adaptation through LoRA on the DINOv3 guide backbone. Across diverse medical datasets and nnUNet-style inference, GuiDINO consistently improves segmentation quality and boundary robustness, suggesting a practical alternative to fine-tuning and offering a new perspective on how foundation models can best serve medical vision. Code is available at https://github.com/Hi-FishU/GuiDINO
Authors: Alp Eren Gen\c{c}o\u{g}lu, Haz{\i}m Kemal Ekenel
Abstract: Reliable post-disaster building damage assessment (BDA) from satellite imagery is hindered by severe class imbalance, background clutter, and domain shift across disaster types and geographies. In this work, we address these problems and explore ways to improve the MambaBDA, the BDA network of ChangeMamba architecture, one of the most successful BDA models. The approach enhances the MambaBDA with three modular components: (i) Focal Loss to mitigate class imbalance damage classification, (ii) lightweight Attention Gates to suppress irrelevant context, and (iii) a compact Alignment Module to spatially warp pre-event features toward post-event content before decoding. We experiment on multiple satellite imagery datasets, including xBD, Pakistan Flooding, Turkey Earthquake, and Ida Hurricane, and conduct in-domain and crossdataset tests. The proposed modular enhancements yield consistent improvements over the baseline model, with 0.8% to 5% performance gains in-domain, and up to 27% on unseen disasters. This indicates that the proposed enhancements are especially beneficial for the generalization capability of the system.
Authors: Xiwei Liu, Yulong Li, Xinlin Zhuang, Xuhui Li, Jianxu Chen, Haolin Yang, Imran Razzak, Yutong Xie
Abstract: Medical Vision-Language Models have shown promising potential in clinical decision support, yet they remain prone to factual hallucinations due to insufficient grounding in localized pathological evidence. Existing medical alignment methods primarily operate at the response level through preference optimization, improving output correctness but leaving intermediate reasoning weakly connected to visual regions. Although chain-of-thought (CoT) enhances multimodal reasoning, it remains largely text-centric, limiting effective integration of clinical visual cues. To address this gap, we propose ClinCoT, a clinical-aware visual chain-of-thought framework that transforms preference optimization from response-level correction to visual-driven reasoning. We introduce an automatic data generation pipeline that constructs clinically grounded preference pairs through reasoning with hypotheses-driven region proposals. Multiple Med-LLMs evaluators rank and assign scores to each response, and these rankings serve as supervision to train the target model. We further introduce a scoring-based margin-aware optimization strategy that incorporates both preference ranking and score difference to refine region-level reasoning trajectories. To maintain alignment as the model's policy evolves during training, we adopt an iterative learning scheme that dynamically regenerates preference data. Extensive experiments on three medical VQA and report generation benchmarks demonstrate that ClinCoT consistently improves factual grounding and achieves superior performance compared with existing preference-based alignment methods.
Authors: Chengtai Li, Yuting He, Jianfeng Ren, Ruibin Bai, Yitian Zhao, Heng Yu, Xudong Jiang
Abstract: While visual reasoning for simple analogies has received significant attention, compositional visual relations (CVR) remain relatively unexplored due to their greater complexity. To solve CVR tasks, we propose Predictive Reasoning with Augmented Anomaly Contrastive Learning (PR-A$^2$CL), \ie, to identify an outlier image given three other images that follow the same compositional rules. To address the challenge of modelling abundant compositional rules, an Augmented Anomaly Contrastive Learning is designed to distil discriminative and generalizable features by maximizing similarity among normal instances while minimizing similarity between normal and anomalous outliers. More importantly, a predict-and-verify paradigm is introduced for rule-based reasoning, in which a series of Predictive Anomaly Reasoning Blocks (PARBs) iteratively leverage features from three out of the four images to predict those of the remaining one. Throughout the subsequent verification stage, the PARBs progressively pinpoint the specific discrepancies attributable to the underlying rules. Experimental results on SVRT, CVR and MC$^2$R datasets show that PR-A$^2$CL significantly outperforms state-of-the-art reasoning models.
Authors: Kuai Jiang, Zhaoyan Ding, Guijuan Zhang, Dianjie Lu, Zhuoran Zheng
Abstract: Conventional image denoising models often inadvertently learn spurious correlations between environmental factors and noise patterns. Moreover, due to high-frequency ambiguity, they struggle to reliably distinguish subtle textures from stochastic noise, resulting in over-removed details or residual noise artifacts. We therefore revisit denoising via causal intervention, arguing that purely correlational fitting entangles intrinsic content with extrinsic noise, which directly degrades robustness under distribution shifts. Motivated by this, we propose the Teacher-Guided Causal Disentanglement Network (TCD-Net), which explicitly decomposes the generative mechanism via structured interventions on feature spaces within a Vision Transformer framework. Specifically, our method integrates three key components: (1) An Environmental Bias Adjustment (EBA) module projects features into a stable, de-centered subspace to suppress global environmental bias (de-confounding). (2) A dual-branch disentanglement head employs an orthogonality constraint to force a strict separation between content and noise representations, preventing information leakage. (3) To resolve structural ambiguity, we leverage Nano Banana Pro, Google's reasoning-guided AI image generation model, to guide a causal prior, effectively pulling content representations back onto the natural-image manifold. Extensive experiments demonstrate that TCD-Net outperforms mainstream methods across multiple benchmarks in both fidelity and efficiency, achieving a real-time speed of 104.2 FPS on a single RTX 5090 GPU.
Authors: Penghao Wang, Siyuan Xie, Hongyu Yan, Xianghui Yang, Jingwei Huang, Chunchao Guo, Jiayuan Gu
Abstract: Creating interactive digital environments for gaming, robotics, and simulation relies on articulated 3D objects whose functionality emerges from their part geometry and kinematic structure. However, existing approaches remain fundamentally limited: optimization-based reconstruction methods require slow, per-object joint fitting and typically handle only simple, single-joint objects, while retrieval-based methods assemble parts from a fixed library, leading to repetitive geometry and poor generalization. To address these challenges, we introduce ArtLLM, a novel framework for generating high-quality articulated assets directly from complete 3D meshes. At its core is a 3D multimodal large language model trained on a large-scale articulation dataset curated from both existing articulation datasets and procedurally generated objects. Unlike prior work, ArtLLM autoregressively predicts a variable number of parts and joints, inferring their kinematic structure in a unified manner from the object's point cloud. This articulation-aware layout then conditions a 3D generative model to synthesize high-fidelity part geometries. Experiments on the PartNet-Mobility dataset show that ArtLLM significantly outperforms state-of-the-art methods in both part layout accuracy and joint prediction, while generalizing robustly to real-world objects. Finally, we demonstrate its utility in constructing digital twins, highlighting its potential for scalable robot learning.
Authors: Zhuo Chen, Shawn Young, Lijian Xu
Abstract: The application of large vision-language models to computational pathology holds great promise for diagnostic assistants but faces a critical computational bottleneck: the gigapixel scale of Whole Slide Images (WSIs). A single WSI typically contains over 105 patches, creating sequence lengths that exceed the constraints of standard Transformer architectures. Existing solutions often resort to spatial sampling, which risks discarding diagnostically critical evidence. To address this, we propose TC-SSA (Token Compression via Semantic Slot Aggregation), a learnable token compression framework that aggregates patch features into a fixed number of semantic slots. A gated routing module assigns patches to slots using sparse Top-2 routing, followed by weighted aggregation, enabling global slide coverage under a strict token budget. The resulting representation retains diagnostically relevant information while reducing the number of visual tokens to 1.7% of the original sequence. On the SlideBench(TCGA), our model achieves 78.34% overall accuracy and 77.14% on the diagnosis subset, outperforming sampling-based baselines under comparable token budgets. The method also generalizes to MIL classification, reaching AUC of 95.83% on TCGA-BRCA, 98.27% on TCGA-NSCLC and 79.80% on PANDA. These results suggest that learnable semantic aggregation provides an effective trade-off between efficiency and diagnostic performance for gigapixel pathology reasoning.
Authors: Jiamei Guo, Zhehao Duan, Maria Neiiendam, Dianye Huang, Nassir Navab, Zhongliang Jiang
Abstract: Purpose: Ultrasound-guided needle interventions are widely used in clinical practice, but their success critically depends on accurate needle placement, which is frequently hindered by the poor and intermittent visibility of needles in ultrasound images. Existing approaches remain limited by artifacts, occlusions, and low contrast, and often fail to support real-time continuous insertion. To overcome these challenges, this study introduces a robust real-time framework for continuous needle detection. Methods: We present ConVibNet, an extension of VibNet for detecting needles with significantly reduced visibility, addressing real-time, continuous needle tracking during insertion. ConVibNet leverages temporal dependencies across successive ultrasound frames to enable continuous estimation of both needle tip position and shaft angle in dynamic scenarios. To strengthen temporal awareness of needle-tip motion, we introduce a novel intersection-and-difference loss that explicitly leverages motion correlations across consecutive frames. In addition, we curated a dedicated dataset for model development and evaluation. Results: The performance of the proposed ConVibNet model was evaluated on our dataset, demonstrating superior accuracy compared to the baseline VibNet and UNet-LSTM models. Specifically, ConVibNet achieved a tip error of 2.80+-2.42 mm and an angle error of 1.69+-2.00 deg. These results represent a 0.75 mm improvement in tip localization accuracy over the best-performing baseline, while preserving real-time inference capability. Conclusion: ConVibNet advances real-time needle detection in ultrasound-guided interventions by integrating temporal correlation modeling with a novel intersection-and-difference loss, thereby improving accuracy and robustness and demonstrating high potential for integration into autonomous insertion systems.
Authors: Durgesh Ameta, Ujjwal Mishra, Praful Hambarde, Amit Shukla
Abstract: Change detection (CD) in remote sensing aims to identify semantic differences between satellite images captured at different times. While deep learning has significantly advanced this field, existing approaches based on convolutional neural networks (CNNs), transformers and Selective State Space Models (SSMs) still struggle to precisely delineate change regions. In particular, traditional transformer-based methods suffer from quadratic computational complexity when applied to very high-resolution (VHR) satellite images and often perform poorly with limited training data, leading to under-utilization of the rich spatial information available in VHR imagery. We present GRAD-Former, a novel framework that enhances contextual understanding while maintaining efficiency through reduced model size. The proposed framework consists of a novel encoder with Adaptive Feature Relevance and Refinement (AFRAR) module, fusion and decoder blocks. AFRAR integrates global-local contextual awareness through two proposed components: the Selective Embedding Amplification (SEA) module and the Global-Local Feature Refinement (GLFR) module. SEA and GLFR leverage gating mechanisms and differential attention, respectively, which generates multiple softmax heaps to capture important features while minimizing the captured irreverent features. Multiple experiments across three challenging CD datasets (LEVIR-CD, CDD, DSIFN-CD) demonstrate GRAD-Former's superior performance compared to existing approaches. Notably, GRAD-Former outperforms the current state-of-the-art models across all the metrics and all the datasets while using fewer parameters. Our framework establishes a new benchmark for remote sensing change detection performance. Our code will be released at: https://github.com/Ujjwal238/GRAD-Former
Authors: Jiachen Yang, Xianhui Lin, Yi Dong, Zebiao Zheng, Xing Liu, Hong Gu, Yanmei Fang
Abstract: Face retouching requires removing subtle imperfections while preserving unique facial identity features, in order to enhance overall aesthetic appeal. However, existing methods suffer from a fundamental trade-off. Supervised learning on labeled data is constrained to pixel-level label mimicry, failing to capture complex subjective human aesthetic preferences. Conversely, while online reinforcement learning (RL) excels at preference alignment, its stochastic exploration paradigm conflicts with the high-fidelity demands of face retouching and often introduces noticeable noise artifacts due to accumulated stochastic drift. To address these limitations, we propose BeautyGRPO, a reinforcement learning framework that aligns face retouching with human aesthetic preferences. We construct FRPref-10K, a fine-grained preference dataset covering five key retouching dimensions, and train a specialized reward model capable of evaluating subtle perceptual differences. To reconcile exploration and fidelity, we introduce Dynamic Path Guidance (DPG). DPG stabilizes the stochastic sampling trajectory by dynamically computing an anchor-based ODE path and replanning a guided trajectory at each sampling timestep, effectively correcting stochastic drift while maintaining controlled exploration. Extensive experiments show that BeautyGRPO outperforms both specialized face retouching methods and general image editing models, achieving superior texture quality, more accurate blemish removal, and overall results that better align with human aesthetic preferences.
Authors: Maomao Li, Yunfei Liu, Yu Li
Abstract: Image-driven video editing aims to propagate edit contents from the modified first frame to the rest frames. The existing methods usually invert the source video to noise using a pre-trained image-to-video (I2V) model and then guide the sampling process using the edited first frame. Generally, a popular choice for maintaining motion and layout from the source video is intervening in the denoising process by injecting attention during reconstruction. However, such injection often leads to unsatisfactory results, where excessive injection leads to conflicting semantics from the source video while insufficient injection brings limited source representation. Recognizing this, we propose an Editing-awaRE (REE) injection method to modulate injection intensity of each token. Specifically, we first compute the pixel difference between the source and edited first frame to form a corresponding editing mask. Next, we track the editing area throughout the entire video by using optical flow to warp the first-frame mask. Then, editing-aware feature injection intensity for each token is generated accordingly, where injection is not conducted on editing areas. Building upon REE injection, we further propose a zero-shot image-driven video editing framework with recent-emerging rectified-Flow models, dubbed FREE-Edit. Without fine-tuning or training, our FREE-Edit demonstrates effectiveness in various image-driven video editing scenarios, showing its capability to produce higher-quality outputs compared with existing techniques. Project page: https://free-edit.github.io/page/.
Authors: Sumin Kim, Hyemin Jeong, Mingu Kang, Yejin Kim, Yoori Oh, Joonseok Lee
Abstract: The exponential growth of video content necessitates effective video summarization to efficiently extract key information from long videos. However, current approaches struggle to fully comprehend complex videos, primarily because they employ static or modality-agnostic fusion strategies. These methods fail to account for the dynamic, frame-dependent variations in modality saliency inherent in video data. To overcome these limitations, we propose TripleSumm, a novel architecture that adaptively weights and fuses the contributions of visual, text, and audio modalities at the frame level. Furthermore, a significant bottleneck for research into multimodal video summarization has been the lack of comprehensive benchmarks. Addressing this bottleneck, we introduce MoSu (Most Replayed Multimodal Video Summarization), the first large-scale benchmark that provides all three modalities. Extensive experiments demonstrate that TripleSumm achieves state-of-the-art performance, outperforming existing methods by a significant margin on four benchmarks, including MoSu. Our code and dataset are available at https://github.com/smkim37/TripleSumm.
Authors: Abdellah Zakaria Sellam, Fadi Abdeladhim Zidi, Salah Eddine Bekhouche, Ihssen Houhou, Marouane Tliba, Cosimo Distante, Abdenour Hadid
Abstract: Accurate classification of hyperspectral imagery (HSI) is often frustrated by the tension between high-dimensional spectral data and the extreme scarcity of labeled training samples. While hierarchical models like LoLA-SpecViT have demonstrated the power of local windowed attention and parameter-efficient fine-tuning, the quadratic complexity of standard Transformers remains a barrier to scaling. We introduce VP-Hype, a framework that rethinks HSI classification by unifying the linear-time efficiency of State-Space Models (SSMs) with the relational modeling of Transformers in a novel hybrid architecture. Building on a robust 3D-CNN spectral front-end, VP-Hype replaces conventional attention blocks with a Hybrid Mamba-Transformer backbone to capture long-range dependencies with significantly reduced computational overhead. Furthermore, we address the label-scarcity problem by integrating dual-modal Visual and Textual Prompts that provide context-aware guidance for the feature extraction process. Our experimental evaluation demonstrates that VP-Hype establishes a new state of the art in low-data regimes. Specifically, with a training sample distribution of only 2\%, the model achieves Overall Accuracy (OA) of 99.69\% on the Salinas dataset and 99.45\% on the Longkou dataset. These results suggest that the convergence of hybrid sequence modeling and multi-modal prompting provides a robust path forward for high-performance, sample-efficient remote sensing.
Authors: Mochu Xiang, Zhelun Shen, Xuesong Li, Jiahui Ren, Jing Zhang, Chen Zhao, Shanshan Liu, Haocheng Feng, Jingdong Wang, Yuchao Dai
Abstract: Human perceive the 3D world through 2D observations from limited viewpoints. While recent feed-forward generalizable 3D reconstruction models excel at recovering 3D structures from sparse images, their representations are often confined to observed regions, leaving unseen geometry un-modeled. This raises a key, fundamental challenge: Can we infer a complete 3D structure from partial 2D observations? We present RnG (Reconstruction and Generation), a novel feed-forward Transformer that unifies these two tasks by predicting an implicit, complete 3D representation. At the core of RnG, we propose a reconstruction-guided causal attention mechanism that separates reconstruction and generation at the attention level, and treats the KV-cache as an implicit 3D representation. Then, arbitrary poses can efficiently query this cache to render high-fidelity, novel-view RGBD outputs. As a result, RnG not only accurately reconstructs visible geometry but also generates plausible, coherent unseen geometry and appearance. Our method achieves state-of-the-art performance in both generalizable 3D reconstruction and novel view generation, while operating efficiently enough for real-time interactive applications. Project page: https://npucvr.github.io/RnG
Authors: Mingkang Dong, Hongyi Cai, Jie Li, Sifan Zhou, Bin Ren, Kunyu Peng, Yuqian Fu
Abstract: The effectiveness of multimodal instruction tuning depends not only on dataset scale, but critically on whether training samples genuinely require visual reasoning. However, existing instruction datasets often contain a substantial portion of visually redundant samples (solvable from text alone), as well as multimodally misaligned supervision that can degrade learning. To address this, we propose VisNec (Visual Necessity Score), a principled data selection framework that measures the marginal contribution of visual input during instruction tuning. By comparing predictive loss with and without visual context, VisNec identifies whether a training instance is vision-critical, redundant, or misaligned. To preserve task diversity, we combine VisNec with semantic clustering and select high-necessity samples within each cluster. Across 10 downstream benchmarks, training on only 15% of the LLaVA-665K dataset selected by VisNec achieves 100.2% of full-data performance. On the smaller Vision-Flan-186K dataset, our selection not only further reduces data size but also surpasses full-data training by 15.8%. These results demonstrate that measuring and leveraging visual necessity provides an effective solution for both efficient and robust multimodal instruction tuning. Codes and selected subsets will be released upon acceptance.
Authors: Li Jin, Weikai Chen, Yujie Wang, Yingda Yin, Zeyu Hu, Runze Zhang, Keyang Luo, Shengju Qian, Xin Wang, Xueying Qin
Abstract: Open-world promptable 3D semantic segmentation remains brittle as semantics are inferred in the input sensor coordinates. Yet, humans, in contrast, interpret parts via functional roles in a canonical space -- wings extend laterally, handles protrude to the side, and legs support from below. Psychophysical evidence shows that we mentally rotate objects into canonical frames to reveal these roles. To fill this gap, we propose \methodName{}, which attains canonical space perception by inducing a latent canonical reference frame learned directly from data. By construction, we create a unified canonical dataset through LLM-guided intra- and cross-category alignment, exposing canonical spatial regularities across 200 categories. By induction, we realize canonicality inside the model through a dual-branch architecture with canonical map anchoring and canonical box calibration, collapsing pose variation and symmetry into a stable canonical embedding. This shift from input pose space to canonical embedding yields far more stable and transferable part semantics. Experimental results show that \methodName{} establishes new state of the art in open-world promptable 3D segmentation.
Authors: Ari Wahl, Dorian Gawlinski, David Przewozny, Paul Chojecki, Felix Bie{\ss}mann, Sebastian Bosse
Abstract: Pre-trained general-purpose Vision-Language Models (VLM) hold the potential to enhance intuitive human-machine interactions due to their rich world knowledge and 2D object detection capabilities. However, VLMs for 3D coordinates detection tasks are rare. In this work, we investigate interactive abilities of VLMs by returning 3D object positions given a monocular RGB image from a wrist-mounted camera, natural language input, and robot states. We collected and curated a heterogeneous dataset of more than 100,000 images and finetuned a VLM using QLoRA with a custom regression head. By implementing conditional routing, our model maintains its ability to process general visual queries while adding specialized 3D position estimation capabilities. Our results demonstrate robust predictive performance with a median MAE of 13 mm on the test set and a five-fold improvement over a simpler baseline without finetuning. In about 25% of the cases, predictions are within a range considered acceptable for the robot to interact with objects.
Authors: Caiyong Piao, Zhiyuan Yan, Haoming Xu, Yunzhen Zhao, Kaiqing Lin, Feiyang Xu, Shuigeng Zhou
Abstract: Accurate rejection of sensitive or harmful visual content, i.e., harmful image guardrail, is critical in many application scenarios. This task must continuously adapt to the evolving safety policies and content across various domains and over time. However, traditional classifiers, confined to fixed categories, require frequent retraining when new policies are introduced. Vision-language models (VLMs) offer a more adaptable and generalizable foundation for dynamic safety guardrails. Despite this potential, existing VLM-based safeguarding methods are typically trained and evaluated under only a fixed safety policy. We find that these models are heavily overfitted to the seen policy, fail to generalize to unseen policies, and even lose the basic instruction-following ability and general knowledge. To address this issue, in this paper we make two key contributions. First, we benchmark the cross-policy generalization performance of existing VLMs with SafeEditBench, a new evaluation suite. SafeEditBench leverages image-editing models to convert unsafe images into safe counterparts, producing policy-aligned datasets where each safe-unsafe image pair remains visually similar except for localized regions violating specific safety rules. Human annotators then provide accurate safe/unsafe labels under five distinct policies, enabling fine-grained assessment of policy-aware generalization. Second, we introduce SafeGuard-VL, a reinforcement learning-based method with verifiable rewards (RLVR) for robust unsafe-image guardrails. Instead of relying solely on supervised fine-tuning (SFT) under fixed policies, SafeGuard-VL explicitly optimizes the model with policy-grounded rewards, promoting verifiable adaptation across evolving policies. Extensive experiments verify the effectiveness of our method for unsafe image guardrails across various policies.
Authors: Changwoo Baek, Jouwon Song, Sohyeon Kim, Kyeongbo Kong
Abstract: Large Vision-Language Models (LVLMs) have adopted visual token pruning strategies to mitigate substantial computational overhead incurred by extensive visual token sequences. While prior works primarily focus on either attention-based or diversity-based pruning methods, in-depth analysis of these approaches' characteristics and limitations remains largely unexplored. In this work, we conduct thorough empirical analysis using effective rank (erank) as a measure of feature diversity and attention score entropy to investigate visual token processing mechanisms and analyze the strengths and weaknesses of each approach. Our analysis reveals two insights: (1) Our erank-based quantitative analysis shows that many diversity-oriented pruning methods preserve substantially less feature diversity than intended; moreover, analysis using the CHAIR dataset reveals that the diversity they do retain is closely tied to increased hallucination frequency compared to attention-based pruning. (2) We further observe that attention-based approaches are more effective on simple images where visual evidence is concentrated, while diversity-based methods better handle complex images with distributed features. Building on these empirical insights, we show that incorporating image-aware adjustments into existing hybrid pruning strategies consistently improves their performance. We also provide a minimal instantiation of our empirical findings through a simple adaptive pruning mechanism, which achieves strong and reliable performance across standard benchmarks as well as hallucination-specific evaluations. Our project page available at https://cvsp-lab.github.io/AgilePruner.
Authors: Lidia Garrucho, Smriti Joshi, Kaisar Kushibar, Richard Osuala, Maciej Bobowicz, Xavier Bargall\'o, Paulius Jaru\v{s}evi\v{c}ius, Kai Geissler, Raphael Sch\"afer, Muhammad Alberb, Tony Xu, Anne Martel, Daniel Sleiman, Navchetan Awasthi, Hadeel Awwad, Joan C. Vilanova, Robert Mart\'i, Daan Schouten, Jeong Hoon Lee, Mirabela Rusu, Eleonora Poeta, Luisa Vargas, Eliana Pastor, Maria A. Zuluaga, Jessica K\"achele, Dimitrios Bounias, Alexandra Ertl, Katarzyna Gwo\'zdziewicz, Maria-Laura Cosaka, Pasant M. Abo-Elhoda, Sara W. Tantawy, Shorouq S. Sakrana, Norhan O. Shawky-Abdelfatah, Amr Muhammad Abdo-Salem, Androniki Kozana, Eugen Divjak, Gordana Ivanac, Katerina Nikiforaki, Michail E. Klontzas, Rosa Garc\'ia-Dosd\'a, Meltem Gulsun-Akpinar, O\u{g}uz Lafc{\i}, Carlos Mart\'in-Isla, Oliver D\'iaz, Laura Igual, Karim Lekadir
Abstract: Breast cancer is the most frequently diagnosed malignancy among women worldwide and a leading cause of cancer-related mortality. Dynamic contrast-enhanced magnetic resonance imaging plays a central role in tumor characterization and treatment monitoring, particularly in patients receiving neoadjuvant chemotherapy. However, existing artificial intelligence models for breast magnetic resonance imaging are often developed using single-center data and evaluated using aggregate performance metrics, limiting their generalizability and obscuring potential performance disparities across demographic subgroups. The MAMA-MIA Challenge was designed to address these limitations by introducing a large-scale benchmark that jointly evaluates primary tumor segmentation and prediction of pathologic complete response using pre-treatment magnetic resonance imaging only. The training cohort comprised 1,506 patients from multiple institutions in the United States, while evaluation was conducted on an external test set of 574 patients from three independent European centers to assess cross-continental and cross-institutional generalization. A unified scoring framework combined predictive performance with subgroup consistency across age, menopausal status, and breast density. Twenty-six international teams participated in the final evaluation phase. Results demonstrate substantial performance variability under external testing and reveal trade-offs between overall accuracy and subgroup fairness. The challenge provides standardized datasets, evaluation protocols, and public resources to promote the development of robust and equitable artificial intelligence systems for breast cancer imaging.
Authors: Timofey Efimov, Singanallur Venkatakrishnan, Maliha Hossain, Haley Duba-Sullivan, Amirkoushyar Ziabari
Abstract: Diffusion models have emerged as powerful priors for solving inverse problems in computed tomography (CT). In certain applications, such as neutron CT, it can be expensive to collect large amounts of measurements even for a single scan, leading to sparse data sets from which it is challenging to obtain high quality reconstructions even with diffusion models. One strategy to mitigate this challenge is to leverage a complementary, easily available imaging modality; however, such approaches typically require retraining the diffusion model with large datasets. In this work, we propose incorporating an additional modality without retraining the diffusion prior, enabling accelerated imaging of costly modalities. We further examine the impact of imperfect side modalities on cross-modal guidance. Our method is evaluated on sparse-view neutron computed tomography, where reconstruction quality is substantially improved by incorporating X-ray computed tomography of the same samples.
Authors: Yizhou Huang, Gengze Jiang, Yihua Cheng, Kezhi Wang
Abstract: Accurate trajectory prediction is vital for safe autonomous driving, yet existing approaches struggle to balance modeling power and computational efficiency. Attention-based architectures incur quadratic complexity with increasing agents, while recurrent models struggle to capture long-range dependencies and fine-grained local dynamics. Building upon this, we present FoSS, a dual-branch framework that unifies frequency-domain reasoning with linear-time sequence modeling. The frequency-domain branch performs a discrete Fourier transform to decompose trajectories into amplitude components encoding global intent and phase components capturing local variations, followed by a progressive helix reordering module that preserves spectral order; two selective state-space submodules, Coarse2Fine-SSM and SpecEvolve-SSM, refine spectral features with O(N) complexity. In parallel, a time-domain dynamic selective SSM reconstructs self-attention behavior in linear time to retain long-range temporal context. A cross-attention layer fuses temporal and spectral representations, while learnable queries generate multiple candidate trajectories, and a weighted fusion head expresses motion uncertainty. Experiments on Argoverse 1 and Argoverse 2 benchmarks demonstrate that FoSS achieves state-of-the-art accuracy while reducing computation by 22.5% and parameters by over 40%. Comprehensive ablations confirm the necessity of each component.
Authors: Abdullah Al Shafi, Md Kawsar Mahmud Khan Zunayed, Safin Ahmmed, Sk Imran Hossain, Engelbert Mephu Nguifo
Abstract: Breast ultrasound interpretation requires simultaneous lesion segmentation and tissue classification. However, conventional multi-task learning approaches suffer from task interference and rigid coordination strategies that fail to adapt to instance-specific prediction difficulty. We propose a multi-task framework addressing these limitations through multi-level decoder interaction and uncertainty-aware adaptive coordination. Task Interaction Modules operate at all decoder levels, establishing bidirectional segmentation-classification communication during spatial reconstruction through attention weighted pooling and multiplicative modulation. Unlike prior single-level or encoder-only approaches, this multi-level design captures scale specific task synergies across semantic-to-spatial scales, producing complementary task interaction streams. Uncertainty-Proxy Attention adaptively weights base versus enhanced features at each level using feature activation variance, enabling per-level and per-sample task balancing without heuristic tuning. To support instance-adaptive prediction, multi-scale context fusion captures morphological cues across varying lesion sizes. Evaluation on multiple publicly available breast ultrasound datasets demonstrates competitive performance, including 74.5% lesion IoU and 90.6% classification accuracy on BUSI dataset. Ablation studies confirm that multi-level task interaction provides significant performance gains, validating that decoder-level bidirectional communication is more effective than conventional encoder-only parameter sharing. The code is available at: https://github.com/C-loud-Nine/Uncertainty-Aware-Multi-Level-Decoder-Interaction.
URLs: https://github.com/C-loud-Nine/Uncertainty-Aware-Multi-Level-Decoder-Interaction.
Authors: Ahmadreza Jeddi, Kimia Shaban, Negin Baghbanzadeh, Natasha Sharan, Abhishek Moturu, Elham Dolatabadi, Babak Taati
Abstract: Reinforcement learning (RL) is increasingly used to post-train medical Vision-Language Models (VLMs), yet it remains unclear whether RL improves medical visual reasoning or mainly sharpens behaviors already induced by supervised fine-tuning (SFT). We present a controlled study that disentangles these effects along three axes: vision, SFT, and RL. Using MedMNIST as a multi-modality testbed, we probe visual perception by benchmarking VLM vision towers against vision-only baselines, quantify reasoning support and sampling efficiency via Accuracy@1 versus Pass@K, and evaluate when RL closes the support gap and how gains transfer across modalities. We find that RL is most effective when the model already has non-trivial support (high Pass@K): it primarily sharpens the output distribution, improving Acc@1 and sampling efficiency, while SFT expands support and makes RL effective. Based on these findings, we propose a boundary-aware recipe and instantiate it by RL post-training an OctoMed-initialized model on a small, balanced subset of PMC multiple-choice VQA, achieving strong average performance across six medical VQA benchmarks.
Authors: Zhen Qu, Xian Tao, Xiaoyi Bao, Dingrong Wang, ShiChen Qu, Zhengtao Zhang, Xingang Wang
Abstract: Large multimodal models (LMMs) exhibit strong task generalization capabilities, offering new opportunities for zero-shot visual anomaly segmentation (ZSAS). However, existing LMM-based segmentation approaches still face fundamental limitations: anomaly concepts are inherently abstract and context-dependent, lacking stable visual prototypes, and the weak alignment between high-level semantic embeddings and pixel-level spatial features hinders precise anomaly localization. To address these challenges, we present AG-VAS (Anchor-Guided Visual Anomaly Segmentation), a new framework that expands the LMM vocabulary with three learnable semantic anchor tokens-[SEG], [NOR], and [ANO], establishing a unified anchor-guided segmentation paradigm. Specifically, [SEG] serves as an absolute semantic anchor that translates abstract anomaly semantics into explicit, spatially grounded visual entities (e.g., holes or scratches), while [NOR] and [ANO] act as relative anchors that model the contextual contrast between normal and abnormal patterns across categories. To further enhance cross-modal alignment, we introduce a Semantic-Pixel Alignment Module (SPAM) that aligns language-level semantic embeddings with high-resolution visual features, along with an Anchor-Guided Mask Decoder (AGMD) that performs anchor-conditioned mask prediction for precise anomaly localization. In addition, we curate Anomaly-Instruct20K, a large-scale instruction dataset that organizes anomaly knowledge into structured descriptions of appearance, shape, and spatial attributes, facilitating effective learning and integration of the proposed semantic anchors. Extensive experiments on six industrial and medical benchmarks demonstrate that AG-VAS achieves consistent state-of-the-art performance in the zero-shot setting.
Authors: Anna Michailidou, Georgios Angelidis, Vasileios Argyriou, Panagiotis Sarigiannidis, Georgios Th. Papadopoulos
Abstract: Aerial imagery is critical for large-scale post-disaster damage assessment. Automated interpretation remains challenging due to clutter, visual variability, and strong cross-event domain shift, while supervised approaches still rely on costly, task-specific annotations with limited coverage across disaster types and regions. Recent open-vocabulary and foundation vision models offer an appealing alternative, by reducing dependence on fixed label sets and extensive task-specific annotations. Instead, they leverage large-scale pretraining and vision-language representations. These properties are particularly relevant for post-disaster domains, where visual concepts are ambiguous and data availability is constrained. In this work, we present a comparative evaluation of supervised learning and open-vocabulary vision models for post-disaster scene understanding, focusing on semantic segmentation and object detection across multiple datasets, including FloodNet+, RescueNet, DFire, and LADD. We examine performance trends, failure modes, and practical trade-offs between different learning paradigms, providing insight into their applicability for real-world disaster response. The most notable remark across all evaluated benchmarks is that supervised training remains the most reliable approach (i.e., when the label space is fixed and annotations are available), especially for small objects and fine boundary delineation in cluttered scenes.
Authors: Taoyue Wang, Xiang Zhang, Xiaotian Li, Huiyuan Yang, Lijun Yin
Abstract: We propose a novel one-stage method, NVB-Face, for generating consistent Novel-View images directly from a single Blind Face image. Existing approaches to novel-view synthesis for objects or faces typically require a high-resolution RGB image as input. When dealing with degraded images, the conventional pipeline follows a two-stage process: first restoring the image to high resolution, then synthesizing novel views from the restored result. However, this approach is highly dependent on the quality of the restored image, often leading to inaccuracies and inconsistencies in the final output. To address this limitation, we extract single-view features directly from the blind face image and introduce a feature manipulator that transforms these features into 3D-aware, multi-view latent representations. Leveraging the powerful generative capacity of a diffusion model, our framework synthesizes high-quality, consistent novel-view face images. Experimental results show that our method significantly outperforms traditional two-stage approaches in both consistency and fidelity.
Authors: Andrew Wang, Mike Davies
Abstract: Multispectral demosaicing is crucial to reconstruct full-resolution spectral images from snapshot mosaiced measurements, enabling real-time imaging from neurosurgery to autonomous driving. Classical methods are blurry, while supervised learning requires costly ground truth (GT) obtained from slow line-scanning systems. We propose Perspective-Equivariant Fine-tuning for Demosaicing (PEFD), a framework that learns multispectral demosaicing from mosaiced measurements alone. PEFD a) exploits the projective geometry of camera-based imaging systems to leverage a richer group structure than previous demosaicing methods to recover more null-space information, and b) learns efficiently without GT by adapting pretrained foundation models designed for 1-3 channel imaging. On intraoperative and automotive datasets, PEFD recovers fine details such as blood vessels and preserves spectral fidelity, substantially outperforming recent approaches, nearing supervised performance.
Authors: Zilong Zhao, Zhengming Ding, Pei Niu, Wenhao Sun, Feng Guo
Abstract: Feature encoders play a key role in pixel-level crack segmentation by shaping the representation of fine textures and thin structures. Existing CNN-, Transformer-, and Mamba-based models each capture only part of the required spatial or structural information, leaving clear gaps in modeling complex crack patterns. To address this, we present MixerCSeg, a mixer architecture designed like a coordinated team of specialists, where CNN-like pathways focus on local textures, Transformer-style paths capture global dependencies, and Mamba-inspired flows model sequential context within a single encoder. At the core of MixerCSeg is the TransMixer, which explores Mamba's latent attention behavior while establishing dedicated pathways that naturally express both locality and global awareness. To further enhance structural fidelity, we introduce a spatial block processing strategy and a Direction-guided Edge Gated Convolution (DEGConv) that strengthens edge sensitivity under irregular crack geometries with minimal computational overhead. A Spatial Refinement Multi-Level Fusion (SRF) module is then employed to refine multi-scale details without increasing complexity. Extensive experiments on multiple crack segmentation benchmarks show that MixerCSeg achieves state-of-the-art performance with only 2.05 GFLOPs and 2.54 M parameters, demonstrating both efficiency and strong representational capability. The code is available at https://github.com/spiderforest/MixerCSeg.
Authors: Xiao Cai, Lianli Gao, Pengpeng Zeng, Ji Zhang, Heng Tao Shen, Jingkuan Song
Abstract: Precise spatial fidelity in Image-to-3D multi-instance generation is critical for downstream real-world applications. Recent work attempts to address this by fine-tuning pre-trained Image-to-3D (I23D) models on multi-instance datasets, which incurs substantial training overhead and struggles to guarantee spatial fidelity. In fact, we observe that pre-trained I23D models already possess meaningful spatial priors, which remain underutilized as evidenced by instance entanglement issues. Motivated by this, we propose TIMI, a novel Training-free framework for Image-to-3D Multi-Instance generation that achieves high spatial fidelity. Specifically, we first introduce an Instance-aware Separation Guidance (ISG) module, which facilitates instance disentanglement during the early denoising stage. Next, to stabilize the guidance introduced by ISG, we devise a Spatial-stabilized Geometry-adaptive Update (SGU) module that promotes the preservation of the geometric characteristics of instances while maintaining their relative relationships. Extensive experiments demonstrate that our method yields better performance in terms of both global layout and distinct local instances compared to existing multi-instance methods, without requiring additional training and with faster inference speed.
Authors: Junwei Zeng, Dong Liang, Sheng-Jun Huang, Kun Zhan, Songcan Chen
Abstract: Atmospheric turbulence significantly degrades long-range imaging by introducing geometric warping and exposure-time-dependent blur, which adversely affects both visual quality and the performance of high-level vision tasks. Existing methods for synthesizing turbulence effects often oversimplify the relationship between blur and exposure-time, typically assuming fixed or binary exposure settings. This leads to unrealistic synthetic data and limited generalization capability of trained models. To address this gap, we revisit the modulation transfer function (MTF) formulation and propose a novel Exposure-Time-dependent MTF (ET-MTF) that models blur as a continuous function of exposure-time. For blur synthesis, we derive a tilt-invariant point spread function (PSF) from the ET-MTF, which, when integrated with a spatially varying blur-width field, provides a comprehensive and physically accurate characterization of turbulence-induced blur. Building on this synthesis pipeline, we construct ET-Turb, a large-scale synthetic turbulence dataset that explicitly incorporates continuous exposure-time modeling across diverse optical and atmospheric conditions. The dataset comprises 5,083 videos (2,005,835 frames), partitioned into 3,988 training and 1,095 test videos. Extensive experiments demonstrate that models trained on ET-Turb produce more realistic restorations and achieve superior generalization on real-world turbulence data compared to those trained on other datasets. The dataset is publicly available at: github.com/Jun-Wei-Zeng/ET-Turb.
Authors: Jinlong Li, Liyuan Jiang, Haonan Zhang, Nicu Sebe
Abstract: Video Large Language Models (VLLMs) demonstrate strong video understanding but suffer from inefficiency due to redundant visual tokens. Existing pruning primary targets intra-frame spatial redundancy or prunes inside the LLM with shallow-layer overhead, yielding suboptimal spatiotemporal reduction and underutilizing long-context compressibility. All of them often discard subtle yet informative context from merged or pruned tokens. In this paper, we propose a new perspective that elaborates token \textbf{A}nchors within intra-frame and inter-frame to comprehensively aggregate the informative contexts via local-global \textbf{O}ptimal \textbf{T}ransport (\textbf{AOT}). Specifically, we first establish local- and global-aware token anchors within each frame under the attention guidance, which then optimal transport aggregates the informative contexts from pruned tokens, constructing intra-frame token anchors. Then, building on the temporal frame clips, the first frame within each clip will be considered as the keyframe anchors to ensemble similar information from consecutive frames through optimal transport, while keeping distinct tokens to represent temporal dynamics, leading to efficient token reduction in a training-free manner. Extensive evaluations show that our proposed AOT obtains competitive performances across various short- and long-video benchmarks on leading video LLMs, obtaining substantial computational efficiency while preserving temporal and visual fidelity. Project webpage: \href{https://tyroneli.github.io/AOT}{AOT}.
Authors: Ben Kang, Jie Zhao, Xin Chen, Wanting Geng, Bin Zhang, Lu Zhang, Dong Wang, Huchuan Lu
Abstract: With growing real-world demands, efficient tracking has received increasing attention. However, most existing methods are limited to RGB inputs and struggle in multi-modal scenarios. Moreover, current multi-modal tracking approaches typically use complex designs, making them too heavy and slow for resource-constrained deployment. To tackle these limitations, we propose UETrack, an efficient framework for single object tracking. UETrack demonstrates high practicality and versatility, efficiently handling multiple modalities including RGB, Depth, Thermal, Event, and Language, and addresses the gap in efficient multi-modal tracking. It introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance. Extensive experiments on 12 benchmarks across 3 hardware platforms show that UETrack achieves a superior speed-accuracy trade-off compared to previous methods. For instance, UETrack-B achieves 69.2% AUC on LaSOT and runs at 163/56/60 FPS on GPU/CPU/AGX, demonstrating strong practicality and versatility. Code is available at https://github.com/kangben258/UETrack.
Authors: Hebeizi Li, Zihao Liang, Benyuan Sun, Zihao Yin, Xiao Sha, Chenliang Wang, Yi Yang
Abstract: While state-of-the-art audio-video generation models like Veo3 and Sora2 demonstrate remarkable capabilities, their closed-source nature makes their architectures and training paradigms inaccessible. To bridge this gap in accessibility and performance, we introduce UniTalking, a unified, end-to-end diffusion framework for generating high-fidelity speech and lip-synchronized video. At its core, our framework employs Multi-Modal Transformer Blocks to explicitly model the fine-grained temporal correspondence between audio and video latent tokens via a shared self-attention mechanism. By leveraging powerful priors from a pre-trained video generation model, our framework ensures state-of-the-art visual fidelity while enabling efficient training. Furthermore, UniTalking incorporates a personalized voice cloning capability, allowing the generation of speech in a target style from a brief audio reference. Qualitative and quantitative results demonstrate that our method produces highly realistic talking portraits, achieving superior performance over existing open-source approaches in lip-sync accuracy, audio naturalness, and overall perceptual quality.
Authors: Yingjian Zhu, Ying Wang, Yuyang Hong, Ruohao Guo, Kun Ding, Xin Gu, Bin Fan, Shiming Xiang
Abstract: Recently, an audio-visual instance segmentation (AVIS) task has been introduced, aiming to identify, segment and track individual sounding instances in videos. However, prevailing methods primarily adopt the offline paradigm, that cannot associate detected instances across consecutive clips, making them unsuitable for real-world scenarios that involve continuous video streams. To address this limitation, we introduce SeaVIS, the first online framework designed for audio-visual instance segmentation. SeaVIS leverages the Causal Cross Attention Fusion (CCAF) module to enable efficient online processing, which integrates visual features from the current frame with the entire audio history under strict causal constraints. A major challenge for conventional VIS methods is that appearance-based instance association fails to distinguish between an object's sounding and silent states, resulting in the incorrect segmentation of silent objects. To tackle this, we employ an Audio-Guided Contrastive Learning (AGCL) strategy to generate instance prototypes that encode not only visual appearance but also sounding activity. In this way, instances preserved during per-frame prediction that do not emit sound can be effectively suppressed during instance association process, thereby significantly enhancing the audio-following capability of SeaVIS. Extensive experiments conducted on the AVISeg dataset demonstrate that SeaVIS surpasses existing state-of-the-art models across multiple evaluation metrics while maintaining a competitive inference speed suitable for real-time processing.
Authors: Zengqi Zhao, Weidi Xia, Peter Wei, Yan Zhang, Yiyi Zhang, Jane Mo, Tiannan Zhang, Yuanqin Dai, Zexi Chen, Simiao Ren
Abstract: We present DOCFORGE-BENCH, the first unified zero-shot benchmark for document forgery detection, evaluating 14 methods across eight datasets spanning text tampering, receipt forgery, and identity document manipulation. Unlike fine-tuning-oriented evaluations such as ForensicHub [Du et al., 2025], DOCFORGE-BENCH applies all methods with their published pretrained weights and no domain adaptation -- a deliberate design choice that reflects the realistic deployment scenario where practitioners lack labeled document training data. Our central finding is a pervasive calibration failure invisible under single-threshold protocols: methods achieve moderate Pixel-AUC (>=0.76) yet near-zero Pixel-F1. This AUC-F1 gap is not a discrimination failure but a score-distribution shift: tampered regions occupy only 0.27-4.17% of pixels in document images -- an order of magnitude less than in natural image benchmarks -- making the standard tau=0.5 threshold catastrophically miscalibrated. Oracle-F1 is 2-10x higher than fixed-threshold Pixel-F1, confirming that calibration, not representation, is the bottleneck. A controlled calibration experiment validates this: adapting a single threshold on N=10 domain images recovers 39-55% of the Oracle-F1 gap, demonstrating that threshold adaptation -- not retraining -- is the key missing step for practical deployment. Overall, no evaluated method works reliably out-of-the-box on diverse document types, underscoring that document forgery detection remains an unsolved problem. We further note that all eight datasets predate the era of generative AI editing; benchmarks covering diffusion- and LLM-based document forgeries represent a critical open gap on the modern attack surface.
Authors: Xinyang Wang, Qian Liu, Wenjie Ding, Zhao Yang, Wei Li, Chang Liu, Bailin Li, Kun Zhan, Xianpeng Lang, Wei Chen
Abstract: Vision-Language-Action (VLA) models are emerging as a promising paradigm for end-to-end autonomous driving, valued for their potential to leverage world knowledge and reason about complex driving scenes. However, existing methods suffer from two critical limitations: a persistent misalignment between language instructions and action outputs, and the inherent inefficiency of typical auto-regressive action generation. In this paper, we introduce LinkVLA, a novel architecture that directly addresses these challenges to enhance both alignment and efficiency. First, we establish a structural link by unifying language and action tokens into a shared discrete codebook, processed within a single multi-modal model. This structurally enforces cross-modal consistency from the ground up. Second, to create a deep semantic link, we introduce an auxiliary action understanding objective that trains the model to generate descriptive captions from trajectories, fostering a bidirectional language-action mapping. Finally, we replace the slow, step-by-step generation with a two-step coarse-to-fine generation method C2F that efficiently decodes the action sequence, saving 86% inference time. Experiments on closed-loop driving benchmarks show consistent gains in instruction following accuracy and driving performance, alongside reduced inference latency.
Authors: Jianfeng Liao, Yichen Wei, Raymond Chan Ching Bon, Shulan Wang, Kam-Pui Chow, Kwok-Yan Lam
Abstract: The rapid advancement of deepfake generation techniques poses significant threats to public safety and causes societal harm through the creation of highly realistic synthetic facial media. While existing detection methods demonstrate limitations in generalizing to emerging forgery patterns, this paper presents Deepfake Forensics Adapter (DFA), a novel dual-stream framework that synergizes vision-language foundation models with targeted forensics analysis. Our approach integrates a pre-trained CLIP model with three core components to achieve specialized deepfake detection by leveraging the powerful general capabilities of CLIP without changing CLIP parameters: 1) A Global Feature Adapter is used to identify global inconsistencies in image content that may indicate forgery, 2) A Local Anomaly Stream enhances the model's ability to perceive local facial forgery cues by explicitly leveraging facial structure priors, and 3) An Interactive Fusion Classifier promotes deep interaction and fusion between global and local features using a transformer encoder. Extensive evaluations of frame-level and video-level benchmarks demonstrate the superior generalization capabilities of DFA, particularly achieving state-of-the-art performance in the challenging DFDC dataset with frame-level AUC/EER of 0.816/0.256 and video-level AUC/EER of 0.836/0.251, representing a 4.8% video AUC improvement over previous methods. Our framework not only demonstrates state-of-the-art performance, but also points out a feasible and effective direction for developing a robust deepfake detection system with enhanced generalization capabilities against the evolving deepfake threats. Our code is available at https://github.com/Liao330/DFA.git
Authors: Duoxun Tang, Dasen Dai, Jiyao Wang, Xiao Yang, Jianyu Wang, Siqi Cai
Abstract: Video-LLMs are increasingly deployed in safety-critical applications but are vulnerable to Energy-Latency Attacks (ELAs) that exhaust computational resources. Current image-centric methods fail because temporal aggregation mechanisms dilute individual frame perturbations. Additionally, real-time demands make instance-wise optimization impractical for continuous video streams. We introduce VidDoS, which is the first universal ELA framework tailored for Video-LLMs. Our method leverages universal optimization to create instance-agnostic triggers that require no inference-time gradient calculation. We achieve this through $\textit{masked teacher forcing}$ to steer models toward expensive target sequences, combined with a $\textit{refusal penalty}$ and $\textit{early-termination suppression}$ to override conciseness priors. Testing across three mainstream Video-LLMs and three video datasets, which include video question answering and autonomous driving scenarios, shows extreme degradation. VidDoS induces a token expansion of more than 205$\times$ and inflates the inference latency by more than 15$\times$ relative to clean baselines. Simulations of real-time autonomous driving streams further reveal that this induced latency leads to critical safety violations. We urge the community to recognize and mitigate these high-hazard ELA in Video-LLMs.
Authors: Niu Lian, Yuting Wang, Hanshu Yao, Jinpeng Wang, Bin Chen, Yaowei Wang, Min Zhang, Shu-Tao Xia
Abstract: While multimodal large language models have demonstrated impressive short-term reasoning, they struggle with long-horizon video understanding due to limited context windows and static memory mechanisms that fail to mirror human cognitive efficiency. Existing paradigms typically fall into two extremes: vision-centric methods that incur high latency and redundancy through dense visual accumulation, or text-centric approaches that suffer from detail loss and hallucination via aggressive captioning. To bridge this gap, we propose MM-Mem, a pyramidal multimodal memory architecture grounded in Fuzzy-Trace Theory. MM-Mem structures memory hierarchically into a Sensory Buffer, Episodic Stream, and Symbolic Schema, enabling the progressive distillation of fine-grained perceptual traces (verbatim) into high-level semantic schemas (gist). Furthermore, to govern the dynamic construction of memory, we derive a Semantic Information Bottleneck objective and introduce SIB-GRPO to optimize the trade-off between memory compression and task-relevant information retention. In inference, we design an entropy-driven top-down memory retrieval strategy, which first tries with the abstract Symbolic Schema and progressively "drills down" to the Sensory Buffer and Episodic Stream under high uncertainty. Extensive experiments across 4 benchmarks confirm the effectiveness of MM-Mem on both offline and streaming tasks, demonstrating robust generalization and validating the effectiveness of cognition-inspired memory organization. Code is available at https://github.com/EliSpectre/MM-Mem.
Authors: Teng Wang, Haojun Jiang, Chenxi Li, Diwen Wang, Yihang Tang, Zhenguo Sun, Yujiao Deng, Shiji Song, Gao Huang
Abstract: Echocardiography is critical for diagnosing cardiovascular diseases, yet the shortage of skilled sonographers hinders timely patient care, due to high operational difficulties. Consequently, research on automated probe navigation has significant clinical potential. To achieve robust navigation, it is essential to leverage historical scanning information, mimicking how experts rely on past feedback to adjust subsequent maneuvers. Practical scanning data collected from sonographers typically consists of noisy trajectories inherently generated through trial-and-error exploration. However, existing methods typically model this history as a sequential chain, forcing models to overfit these noisy paths, leading to performance degradation on long sequences. In this paper, we propose UltraStar, which reformulates probe navigation from path regression to anchor-based global localization. By establishing a Star Graph, UltraStar treats historical keyframes as spatial anchors connected directly to the current view, explicitly modeling geometric constraints for precise positioning. We further enhance the Star Graph with a semantic-aware sampling strategy that actively selects the representative landmarks from massive history logs, reducing redundancy for accurate anchoring. Extensive experiments on a dataset with over 1.31 million samples demonstrate that UltraStar outperforms baselines and scales better with longer input lengths, revealing a more effective topology for history modeling under noisy exploration.
Authors: Joshua Knights, Joseph Reid, Kaushik Roy, David Hall, Mark Cox, Peyman Moghadam
Abstract: Recent years have seen a significant increase in demand for robotic solutions in unstructured natural environments, alongside growing interest in bridging 2D and 3D scene understanding. However, existing robotics datasets are predominantly captured in structured urban environments, making them inadequate for addressing the challenges posed by complex, unstructured natural settings. To address this gap, we propose WildCross, a cross-modal benchmark for place recognition and metric depth estimation in large-scale natural environments. WildCross comprises over 476K sequential RGB frames with semi-dense depth and surface normal annotations, each aligned with accurate 6DoF poses and synchronized dense lidar submaps. We conduct comprehensive experiments on visual, lidar, and cross-modal place recognition, as well as metric depth estimation, demonstrating the value of WildCross as a challenging benchmark for multi-modal robotic perception tasks. We provide access to the code repository and dataset at https://csiro-robotics.github.io/WildCross.
Authors: Brian Cheong, Letian Wang, Sandro Papais, Steven L. Waslander
Abstract: LiDAR-based tracking-by-attention (TBA) frameworks inherently suffer from high false negative errors, leading to a significant performance gap compared to traditional LiDAR-based tracking-by-detection (TBD) methods. This paper introduces SCATR, a novel LiDAR-based TBA model designed to address this fundamental challenge systematically. SCATR leverages recent progress in vision-based tracking and incorporates targeted training strategies specifically adapted for LiDAR. Our work's core innovations are two architecture-agnostic training strategies for TBA methods: Second Chance Assignment and Track Query Dropout. Second Chance Assignment is a novel ground truth assignment that concatenates unassigned track queries to the proposal queries before bipartite matching, giving these track queries a second chance to be assigned to a ground truth object and effectively mitigating the conflict between detection and tracking tasks inherent in tracking-by-attention. Track Query Dropout is a training method that diversifies supervised object query configurations to efficiently train the decoder to handle different track query sets, enhancing robustness to missing or newborn tracks. Experiments on the nuScenes tracking benchmark demonstrate that SCATR achieves state-of-the-art performance among LiDAR-based TBA methods, outperforming previous works by 7.6\% AMOTA and successfully bridging the long-standing performance gap between LiDAR-based TBA and TBD methods. Ablation studies further validate the effectiveness and generalization of Second Chance Assignment and Track Query Dropout. Code can be found at the following link: \href{https://github.com/TRAILab/SCATR}{https://github.com/TRAILab/SCATR}
URLs: https://github.com/TRAILab/SCATR, https://github.com/TRAILab/SCATR
Authors: Cheng Yang, Jianhao Jiao, Lingyi Huang, Jinqi Xiao, Zhexiang Tang, Yu Gong, Yibiao Ying, Yang Sui, Jintian Lin, Wen Huang, Bo Yuan
Abstract: Vision-Language-Action (VLA) models rely on current observations, including images, language instructions, and robot states, to predict actions and complete tasks. While accurate visual perception is crucial for precise action prediction and execution, recent work has attempted to further improve performance by introducing explicit reasoning during inference. However, such approaches face significant limitations. They often depend on data-intensive resources such as Chain-of-Thought (CoT) style annotations to decompose tasks into step-by-step reasoning, and in many cases require additional visual grounding annotations (e.g., bounding boxes or masks) to highlight relevant image regions. Moreover, they involve time-consuming dataset construction, labeling, and retraining, which ultimately results in longer inference sequences and reduced efficiency. To address these challenges, we propose ATA, a novel training-free framework that introduces implicit reasoning into VLA inference through complementary attention-guided and action-guided strategies. Unlike CoT or explicit visual-grounding methods, ATA formulates reasoning implicitly by integrating attention maps with an action-based region of interest (RoI), thereby adaptively refining visual inputs without requiring extra training or annotations. ATA is a plug-and-play implicit reasoning approach for VLA models, lightweight yet effective. Extensive experiments show that it consistently improves task success and robustness while preserving, and even enhancing, inference efficiency.
Authors: Kyu Beom Han, Jaeyoon Kim, Woo Jae Kim, Jinhwan Seo, Sung-eui Yoon
Abstract: Inverse rendering with Gaussian Splatting has advanced rapidly, but accurately disentangling material properties from complex global illumination effects, particularly indirect illumination, remains a major challenge. Existing methods often query indirect radiance from Gaussian primitives pre-trained for novel-view synthesis. However, these pre-trained Gaussian primitives are supervised only towards limited training viewpoints, thus lack supervision for modeling indirect radiances from unobserved views. To address this issue, we introduce radiometric consistency, a novel physically-based constraint that provides supervision towards unobserved views by minimizing the residual between each Gaussian primitive's learned radiance and its physically-based rendered counterpart. Minimizing the residual for unobserved views establishes a self-correcting feedback loop that provides supervision from both physically-based rendering and novel-view synthesis, enabling accurate modeling of inter-reflection. We then propose Radiometrically Consistent Gaussian Surfels (RadioGS), an inverse rendering framework built upon our principle by efficiently integrating radiometric consistency by utilizing Gaussian surfels and 2D Gaussian ray tracing. We further propose a finetuning-based relighting strategy that adapts Gaussian surfel radiances to new illuminations within minutes, achieving low rendering cost (<10ms). Extensive experiments on existing inverse rendering benchmarks show that RadioGS outperforms existing Gaussian-based methods in inverse rendering, while retaining the computational efficiency.
Authors: Kai Zheng, Hang-Cheng Dong, Zhenkai Wu, Fupeng Wei, Wei Zhang
Abstract: In remote sensing imagery, multi class change detection (MCD) is crucial for fine grained monitoring, yet it has long been constrained by complex scene variations and the scarcity of detailed annotations. To address this, we propose the Tripath DINO architecture, which adopts a three path complementary feature learning strategy to facilitate the rapid adaptation of pre trained foundation models to complex vertical domains. Specifically, we employ the DINOv3 pre trained model as the backbone feature extraction network to learn coarse grained features. An auxiliary path also adopts a siamese structure, progressively aggregating intermediate features from the siamese encoder to enhance the learning of fine grained features. Finally, a multi scale attention mechanism is introduced to augment the decoder network, where parallel convolutions adaptively capture and enhance contextual information under different receptive fields. The proposed method achieves optimal performance on the MCD task on both the Gaza facility damage assessment dataset (Gaza change) and the classic SECOND dataset. GradCAM visualizations further confirm that the main and auxiliary paths naturally focus on coarse grained semantic changes and fine grained structural details, respectively. This synergistic complementarity provides a robust and interpretable solution for advanced change detection tasks, offering a basis for rapid and accurate damage assessment.
Authors: Jianqiang Ren, Lin Liu, Steven Hoi
Abstract: We propose OMG-Avatar, a novel One-shot method that leverages a Multi-LOD (Level-of-Detail) Gaussian representation for animatable 3D head reconstruction from a single image in 0.2s. Our method enables LOD head avatar modeling using a unified model that accommodates diverse hardware capabilities and inference speed requirements. To capture both global and local facial characteristics, we employ a transformer-based architecture for global feature extraction and projection-based sampling for local feature acquisition. These features are effectively fused under the guidance of a depth buffer, ensuring occlusion plausibility. We further introduce a coarse-to-fine learning paradigm to support Level-of-Detail functionality and enhance the perception of hierarchical details. To address the limitations of 3DMMs in modeling non-head regions such as the shoulders, we introduce a multi-region decomposition scheme in which the head and shoulders are predicted separately and then integrated through cross-region combination. Extensive experiments demonstrate that OMG-Avatar outperforms state-of-the-art methods in reconstruction quality, reenactment performance, and computational efficiency.
Authors: Zillur Rahman, Alex Sheng, Cristian Meo
Abstract: While large-scale datasets have driven significant progress in Text-to-Video (T2V) generative models, these models remain highly sensitive to input prompts, demonstrating that prompt design is critical to generation quality. Current methods for improving video output often fall short: they either depend on complex, post-editing models, risking the introduction of artifacts, or require expensive fine-tuning of the core generator, which severely limits both scalability and accessibility. In this work, we introduce 3R, a novel RAG based prompt optimization framework. 3R utilizes the power of current state-of-the-art T2V diffusion model and vision language model. It can be used with any T2V model without any kind of model training. The framework leverages three key strategies: RAG-based modifiers extraction for enriched contextual grounding, diffusion-based Preference Optimization for aligning outputs with human preferences, and temporal frame interpolation for producing temporally consistent visual contents. Together, these components enable more accurate, efficient, and contextually aligned text-to-video generation. Experimental results demonstrate the efficacy of 3R in enhancing the static fidelity and dynamic coherence of generated videos, underscoring the importance of optimizing user prompts.
Authors: Hanxiao Wang, Yuan-Chen Guo, Ying-Tian Liu, Zi-Xin Zou, Biao Zhang, Weize Quan, Ding Liang, Yan-Pei Cao, Dong-Ming Yan
Abstract: Autoregressive models for 3D mesh generation suffer from a fundamental limitation: they flatten meshes into long vertex-coordinate sequences. This results in prohibitive computational costs, hindering the efficient synthesis of high-fidelity geometry. We argue this bottleneck stems from operating at the wrong semantic level. We introduce FACE, a novel Autoregressive Autoencoder (ARAE) framework that reconceptualizes the task by generating meshes at the face level. Our one-face-one-token strategy treats each triangle face, the fundamental building block of a mesh, as a single, unified token. This simple yet powerful design reduces the sequence length by a factor of nine, leading to an unprecedented compression ratio of 0.11, halving the previous state-of-the-art. This dramatic efficiency gain does not compromise quality; by pairing our face-level decoder with a powerful VecSet encoder, FACE achieves state-of-the-art reconstruction quality on standard benchmarks. The versatility of the learned latent space is further demonstrated by training a latent diffusion model that achieves high-fidelity, single-image-to-mesh generation. FACE provides a simple, scalable, and powerful paradigm that lowers the barrier to high-quality structured 3D content creation.
Authors: Qirui Wu, Shizhou Zhang, De Cheng, Yinghui Xing, Lingyan Ran, Dahu Shi, Peng Wang
Abstract: Incremental Object Detection (IOD) aims to continuously learn new object classes without forgetting previously learned ones. A persistent challenge is catastrophic forgetting, primarily attributed to background shift in conventional detectors. While pseudo-labeling mitigates this in dense detectors, we identify a novel, distinct source of forgetting specific to DETR-like architectures: background foregrounding. This arises from the exhaustiveness constraint of the Hungarian matcher, which forcibly assigns every ground truth target to one prediction, even when predictions primarily cover background regions (i.e., low IoU). This erroneous supervision compels the model to misclassify background features as specific foreground classes, disrupting learned representations and accelerating forgetting. To address this, we propose a Quality-guided Min-Cost Max-Flow (Q-MCMF) matcher. To avoid forced assignments, Q-MCMF builds a flow graph and prunes implausible matches based on geometric quality. It then optimizes for the final matching that minimizes cost and maximizes valid assignments. This strategy eliminates harmful supervision from background foregrounding while maximizing foreground learning signals. Extensive experiments on the COCO dataset under various incremental settings demonstrate that our method consistently outperforms existing state-of-the-art approaches.
Authors: Yutian Zhang, Zhongyi Pei, Yi Mao, Chen Wang, Lin Liu, Jianmin Wang
Abstract: The widespread adoption of AI in industry is often hampered by its limited robustness when faced with scenarios absent from training data, leading to prediction bias and vulnerabilities. To address this, we propose a novel streaming inference pipeline that enhances data-driven models by explicitly incorporating prior knowledge. This paper presents the work on an industrial AI application that automatically counts excavator workloads from surveillance videos. Our approach integrates an object detection model with a Finite State Machine (FSM), which encodes knowledge of operational scenarios to guide and correct the AI's predictions on streaming data. In experiments on a real-world dataset of over 7,000 images from 12 site videos, encompassing more than 300 excavator workloads, our method demonstrates superior performance and greater robustness compared to the original solution based on manual heuristic rules. We will release the code at https://github.com/thulab/video-streamling-inference-pipeline.
URLs: https://github.com/thulab/video-streamling-inference-pipeline.
Authors: Zijin Yin, Bing Li, Kongming Liang, Hao Sun, Zhongjiang He, Zhanyu Ma, Jun Guo
Abstract: Semantic segmentation takes pivotal roles in various applications such as autonomous driving and medical image analysis. When deploying segmentation models in practice, it is critical to test their behaviors in varied and complex scenes in advance. In this paper, we construct an automatic data generation pipeline Gen4Seg to stress-test semantic segmentation models by generating various challenging samples with different attribute changes. Beyond previous evaluation paradigms focusing solely on global weather and style transfer, we investigate variations in both appearance and geometry attributes at the object and image level. These include object color, material, size, position, as well as image-level variations such as weather and style. To achieve this, we propose to edit visual attributes of existing real images with precise control of structural information, empowered by diffusion models. In this way, the existing segmentation labels can be reused for the edited images, which greatly reduces the labor costs. Using our pipeline, we construct two new benchmarks, Pascal-EA and COCO-EA. We benchmark a wide variety of semantic segmentation models, spanning from closed-set models to open-vocabulary large models. We have several key findings: 1) advanced open-vocabulary models do not exhibit greater robustness compared to closed-set methods under geometric variations; 2) data augmentation techniques, such as CutOut and CutMix, are limited in enhancing robustness against appearance variations; 3) our pipeline can also be employed as a data augmentation tool and improve both in-distribution and out-of-distribution performances. Our work suggests the potential of generative models as effective tools for automatically analyzing segmentation models, and we hope our findings will assist practitioners and researchers in developing more robust and reliable segmentation models.
Authors: Xinchang Wang, Yunhao Chen, Yuechen Zhang, Congcong Bian, Zihao Guo, Xingjun Ma, Hui Li
Abstract: Recent image generators produce photo-realistic content that undermines the reliability of downstream recognition systems. As visual appearance cues become less pronounced, appearance-driven detectors that rely on forensic cues or high-level representations lose stability. This motivates a shift from appearance to behavior, focusing on how images respond to controlled perturbations rather than how they look. In this work, we identify a simple and universal behavioral signal. Natural images preserve stable semantic representations under small, structured perturbations, whereas generated images exhibit markedly larger feature drift. We refer to this phenomenon as robustness asymmetry and provide a theoretical analysis that establishes a lower bound connecting this asymmetry to memorization tendencies in generative models, explaining its prevalence across architectures. Building on this insight, we introduce Robustness Asymmetry Detection (RA-Det), a behavior-driven detection framework that converts robustness asymmetry into a reliable decision signal. Evaluated across 14 diverse generative models and against more than 10 strong detectors, RA-Det achieves superior performance, improving the average performance by 7.81 percent. The method is data- and model-agnostic, requires no generator fingerprints, and transfers across unseen generators. Together, these results indicate that robustness asymmetry is a stable, general cue for synthetic-image detection and that carefully designed probing can turn this cue into a practical, universal detector. The source code is publicly available at Github.
Authors: Zhengtong Zhu, Jiaqing Fan, Zhixuan Liu, Fanzhang Li
Abstract: Reasoning Video Object Segmentation (ReasonVOS) is a challenging task that requires stable object segmentation across video sequences using implicit and complex textual inputs. Previous methods fine-tune Multimodal Large Language Models (MLLMs) to produce segmentation outputs, which demand substantial resources. Additionally, some existing methods are coupled in the processing of spatio-temporal information, which affects the temporal stability of the model to some extent. To address these issues, we propose Training-Free \textbf{S}patio-temporal \textbf{D}ecoupled Reasoning Video Segmentation with \textbf{A}daptive Object \textbf{M}emory (SDAM). We aim to design a training-free reasoning video segmentation framework that outperforms existing methods requiring fine-tuning, using only pre-trained models. Meanwhile, we propose an Adaptive Object Memory module that selects and memorizes key objects based on motion cues in different video sequences. Finally, we propose Spatio-temporal Decoupling for stable temporal propagation. In the spatial domain, we achieve precise localization and segmentation of target objects, while in the temporal domain, we leverage key object temporal information to drive stable cross-frame propagation. Our method achieves excellent results on five benchmark datasets, including Ref-YouTubeVOS, Ref-DAVIS17, MeViS, ReasonVOS, and ReVOS.
Authors: Jian Yu, Joakim Nguyen, Jinrui Fang, Awais Naeem, Zeyuan Cao, Sanjay Krishnan, Nicholas Konz, Tianlong Chen, Chandra Krishnan, Hairong Wang, Edward Castillo, Ying Ding, Ankita Shukla
Abstract: Accurate classification of pediatric central nervous system tumors remains challenging due to histological complexity and limited training data. While pathology foundation models have advanced whole-slide image (WSI) analysis, they often fail to leverage the rich, complementary information found in clinical text and tissue microarchitecture. To this end, we propose PathMoE, an interpretable multimodal framework that integrates H\&E slides, pathology reports, and nuclei-level cell graphs via an interaction-aware mixture-of-experts architecture built on state-of-the-art foundation models for each modality. By training specialized experts to capture modality uniqueness, redundancy, and synergy, PathMoE employs an input-dependent gating mechanism that dynamically weights these interactions, providing sample-level interpretability. We evaluate our framework on two dataset-specific classification tasks on an internal pediatric brain tumor dataset (PBT) and external TCGA datasets. PathMoE improves macro-F1 from 0.762 to 0.799 (+0.037) on PBT when integrating WSI, text, and graph modalities; on TCGA, augmenting WSI with graph knowledge improves macro-F1 from 0.668 to 0.709 (+0.041). These results demonstrate significant performance gains over state-of-the-art image-only baselines while revealing the specific modality interactions driving individual predictions. This interpretability is particularly critical for rare tumor subtypes, where transparent model reasoning is essential for clinical trust and diagnostic validation.
Authors: Jisoo Kim, Jungbin Cho, Sanghyeok Chu, Ananya Bal, Jinhyung Kim, Gunhee Lee, Sihaeng Lee, Seung Hwan Kim, Bohyung Han, Hyunmin Lee, Laszlo A. Jeni, Seungryong Kim
Abstract: Humans learn not only how their bodies move, but also how the surrounding world responds to their actions. In contrast, while recent Vision-Language-Action (VLA) models exhibit impressive semantic understanding, they often fail to capture the spatiotemporal dynamics governing physical interaction. In this paper, we introduce Pri4R, a simple yet effective approach that endows VLA models with an implicit understanding of world dynamics by leveraging privileged 4D information during training. Specifically, Pri4R augments VLAs with a lightweight point track head that predicts 3D point tracks. By injecting VLA features into this head to jointly predict future 3D trajectories, the model learns to incorporate evolving scene geometry within its shared representation space, enabling more physically aware context for precise control. Due to its architectural simplicity, Pri4R is compatible with dominant VLA design patterns with minimal changes. During inference, we run the model using the original VLA architecture unchanged; Pri4R adds no extra inputs, outputs, or computational overhead. Across simulation and real-world evaluations, Pri4R significantly improves performance on challenging manipulation tasks, including a +10% gain on LIBERO-Long and a +40% gain on RoboCasa. We further show that 3D point track prediction is an effective supervision target for learning action-world dynamics, and validate our design choices through extensive ablations.
Authors: Ayantika Das, Keerthi Ram, Mohanasankar Sivaprakasam
Abstract: Generative AI framework-based modeling and prediction of longitudinal human brain images offer an efficient mechanism to track neurodegenerative progression, essential for the assessment of diseases like Alzheimer's. Among the existing generative approaches, recent diffusion-based models have emerged as an effective alternative to generate disease progression images. Incorporating multi-modal and non-imaging attributes as conditional information into diffusion frameworks has been shown to improve controllability during such generations. However, existing methods do not explicitly ensure that information from non-imaging conditioning modalities is meaningfully aligned with image features to introduce desirable changes in the generated images, such as modulation of progression-specific regions. Further, more precise control over the generation process can be achieved by introducing progression-relevant structure into the internal representations of the model, lacking in the existing approaches. To address these limitations, we propose a diffusion autoencoder-based framework for disease progression modeling that explicitly enforces alignment between different modalities. The alignment is enforced by introducing an explicit objective function that enables the model to focus on the regions exhibiting progression-related changes. Further, we devise a mechanism to better structure the latent representational space of the diffusion auto-encoding framework. Specifically, we assign separate latent subspaces for integrating progression-related conditions and retaining subject-specific identity information, allowing better-controlled image generation. These results demonstrate that enforcing alignment and better structuring of the latent representational space of diffusion auto-encoding framework leads to more anatomically precise modeling of Alzheimer's disease progression.
Authors: Muhammet Esat Kalfaoglu, Halil Ibrahim Ozturk, Ozsel Kilinc, Alptekin Temizel
Abstract: Mask-based paradigms for road topology understanding, such as TopoMaskV2, offer a complementary alternative to query-based methods by generating centerlines via a dense rasterized intermediate representation. However, prior work was limited to 2D predictions and suffered from severe discretization artifacts, necessitating fusion with parametric heads. We introduce TopoMaskV3, which advances this pipeline into a robust, standalone 3D predictor via two novel dense prediction heads: a dense offset field for sub-grid discretization correction within the existing BEV resolution, and a dense height map for direct 3D estimation. Beyond the architecture, we are the first to address geographic data leakage in road topology evaluation by introducing (1) geographically distinct splits to prevent memorization and ensure fair generalization, and (2) a long-range (+/-100 m) benchmark. TopoMaskV3 achieves state-of-the-art 28.5 OLS on this geographically disjoint benchmark, surpassing all prior methods. Our analysis shows that the mask representation is more robust to geographic overfitting than Bezier, while LiDAR fusion is most beneficial at long range and exhibits larger relative gains on the overlapping original split, suggesting overlap-induced memorization effects.
Authors: Saurabh Kaushik, Lalit Maurya, Beth Tellman
Abstract: Geo-Foundation Models (GFMs) have been evaluated across diverse Earth observation task including multiple domains and have demonstrated strong potential of producing reliable maps even with sparse labels. However, benchmarking GFMs for Cryosphere applications has remained limited, primarily due to the lack of suitable evaluation datasets. To address this gap, we introduce \textbf{Cryo-Bench}, a benchmark compiled to evaluate GFM performance across key Cryospheric components. Cryo-Bench includes debris-covered glaciers, glacial lakes, sea ice, and calving fronts, spanning multiple sensors and broad geographic regions. We evaluate 14 GFMs alongside UNet and ViT baselines to assess their advantages, limitations, and optimal usage strategies. With a frozen encoder, UNet achieves the highest average mIoU of \textbf{66.38}, followed by TerraMind at \textbf{64.02} across five evluation dataset included in Cryo-Bench. In the few-shot setting (10\% input data), GFMs such as DOFA and TerraMind outperform UNet, achieving mIoU scores of \textbf{59.53}, \textbf{56.62}, and \textbf{56.60}, respectively, comapred to U-Net's 56.60. When fully finetuning GFMs, we observe inconsistent performance across datasets and models. However, tuning learning rate along with finetuning substantially improves GFM performance. For example, evaluation on two representative datasets (GLID and CaFFe) shows an average relative improvement of \textbf{12.77\%}. Despite having minimal Cryosphere representation in their pretraining data, GFMs exhibit notable domain adaptation capabilities and produce meaningful results across tasks. Based on our findings, We recommend encoder fine-tuning with hyperparameter optimization optimization to achieve the best possible performance, while using frozen encoders when users need quick results without extensive experimentation.(\href{https://github.com/Sk-2103/Cryo-Bench}{GitHub}).
Authors: Chuqiao Wu, Jin Song, Yiyun Fei
Abstract: Generating realistic and structurally plausible human images into existing scenes remains a significant challenge for current generative models, which often produce artifacts like distorted limbs and unnatural poses. We attribute this systemic failure to an inability to perform explicit reasoning over human skeletal structure. To address this, we introduce SkeleGuide, a novel framework built upon explicit skeletal reasoning. Through joint training of its reasoning and rendering stages, SkeleGuide learns to produce an internal pose that acts as a strong structural prior, guiding the synthesis towards high structural integrity. For fine-grained user control, we introduce PoseInverter, a module that decodes this internal latent pose into an explicit and editable format. Extensive experiments demonstrate that SkeleGuide significantly outperforms both specialized and general-purpose models in generating high-fidelity, contextually-aware human images. Our work provides compelling evidence that explicitly modeling skeletal structure is a fundamental step towards robust and plausible human image synthesis.
Authors: Yecong Wan, Fan Li, Chunwei Wang, Hao Wu, Mingwen Shao, Wangmeng Zuo
Abstract: Emerging unified editing models have demonstrated strong capabilities in general object editing tasks. However, it remains a significant challenge to perform fine-grained editing in complex multi-entity scenes, particularly those where targets are not visually salient and require spatial reasoning. To this end, we propose InterCoG, a novel text-vision Interleaved Chain-of-Grounding reasoning framework for fine-grained image editing in complex real-world scenes. The key insight of InterCoG is to first perform object position reasoning solely within text that includes spatial relation details to explicitly deduce the location and identity of the edited target. It then conducts visual grounding via highlighting the editing targets with generated bounding boxes and masks in pixel space, and finally rewrites the editing description to specify the intended outcomes. To further facilitate this paradigm, we propose two auxiliary training modules: multimodal grounding reconstruction supervision and multimodal grounding reasoning alignment to enforce spatial localization accuracy and reasoning interpretability, respectively. We also construct GroundEdit-45K, a dataset comprising 45K grounding-oriented editing samples with detailed reasoning annotations, and GroundEdit-Bench for grounding-aware editing evaluation. Extensive experiments substantiate the superiority of our approach in highly precise edits under spatially intricate and multi-entity scenes.
Authors: Bo Ma, Jinsong Wu, Weiqi Yan, Catherine Shi, Minh Nguyen
Abstract: Dashcam videos collected by autonomous or assisted-driving systems are increasingly shared for safety auditing and model improvement. Even when explicit GPS metadata are removed, an attacker can still infer the recording location by matching background visual cues (e.g., buildings and road layouts) against large-scale street-view imagery. This paper studies location-privacy leakage under a background-based retrieval attacker, and proposes PPEDCRF, a privacy-preserving enhanced dynamic conditional random field framework that injects calibrated perturbations only into inferred location-sensitive background regions while preserving foreground detection utility. PPEDCRF consists of three components: (i) a dynamic CRF that enforces temporal consistency to discover and track location sensitive regions across frames, (ii) a normalized control penalty (NCP) that allocates perturbation strength according to a hierarchical sensitivity model, and (iii) a utility-preserving noise injection module that minimizes interference to object detection and segmentation. Experiments on public driving datasets demonstrate that PPEDCRF significantly reduces location-retrieval attack success (e.g., Top-k retrieval accuracy) while maintaining competitive detection performance (e.g., mAP and segmentation metrics) compared with common baselines such as global noise, white-noise masking, and feature-based anonymization. The source code is in https://github.com/mabo1215/PPEDCRF.git
Authors: Jiaqi Leng, Shuyuan Tu, Haidong Cao, Sicheng Xie, Daoguo Dong, Zuxuan Wu, Yu-Gang Jiang
Abstract: Human preference alignment presents a critical yet underexplored challenge for diffusion models in text-to-3D generation. Existing solutions typically require task-specific fine-tuning, posing significant hurdles in data-scarce 3D domains. To address this, we propose Preference Score Distillation (PSD), an optimization-based framework that leverages pretrained 2D reward models for human-aligned text-to-3D synthesis without 3D training data. Our key insight stems from the incompatibility of pixel-level gradients: due to the absence of noisy samples during reward model training, direct application of 2D reward gradients disturbs the denoising process. Noticing that similar issue occurs in the naive classifier guidance in conditioned diffusion models, we fundamentally rethink preference alignment as a classifier-free guidance (CFG)-style mechanism through our implicit reward model. Furthermore, recognizing that frozen pretrained diffusion models constrain performance, we introduce an adaptive strategy to co-optimize preference scores and negative text embeddings. By incorporating CFG during optimization, online refinement of negative text embeddings dynamically enhances alignment. To our knowledge, we are the first to bridge human preference alignment with CFG theory under score distillation framework. Experiments demonstrate the superiority of PSD in aesthetic metrics, seamless integration with diverse pipelines, and strong extensibility.
Authors: Xiwen Wang, Shichao Zhang, Hailun Zhang, Ruowei Wang, Mao Li, Chenyu Zhou, Qijun Zhao, Ji-Zhe Zhou
Abstract: Large 3D reconstruction models have revolutionized the 3D content generation field, enabling broad applications in virtual reality and gaming. Just like other large models, large 3D reconstruction models suffer from hallucinations as well, introducing structural outliers (e.g., odd holes or protrusions) that deviate from the input data. However, unlike other large models, hallucinations in large 3D reconstruction models remain severely underexplored, leading to malformed 3D-printed objects or insufficient immersion in virtual scenes. Such hallucinations majorly originate from that existing methods reconstruct 3D content from sparsely generated multi-view images which suffer from large viewpoint gaps and discontinuities. To mitigate hallucinations by eliminating the outliers, we propose Dehallu3D for 3D mesh generation. Our key idea is to design a balanced multi-view continuity constraint to enforce smooth transitions across dense intermediate viewpoints, while avoiding over-smoothing that could erase sharp geometric features. Therefore, Dehallu3D employs a plug-and-play optimization module with two key constraints: (i) adjacent consistency to ensure geometric continuity across views, and (ii) adaptive smoothness to retain fine details.We further propose the Outlier Risk Measure (ORM) metric to quantify geometric fidelity in 3D generation from the perspective of outliers. Extensive experiments show that Dehallu3D achieves high-fidelity 3D generation by effectively preserving structural details while removing hallucinated outliers.
Authors: PeiHuang Zheng, Yunlong Zhao, Zheng Cui, Yang Li
Abstract: Human vision exhibits remarkable adaptability in perceiving objects under camouflage. When color cues become unreliable, the visual system instinctively shifts its reliance from chrominance (color) to luminance (brightness and texture), enabling more robust perception in visually confusing environments. Drawing inspiration from this biological mechanism, we propose YCDa, an efficient early-stage feature processing strategy that embeds this "chrominance-luminance decoupling and dynamic attention" principle into modern real-time detectors. Specifically, YCDa separates color and luminance information in the input stage and dynamically allocates attention across channels to amplify discriminative cues while suppressing misleading color noise. The strategy is plug-and-play and can be integrated into existing detectors by simply replacing the first downsampling layer. Extensive experiments on multiple baselines demonstrate that YCDa consistently improves performance with negligible overhead as shown in Fig. Notably, YCDa-YOLO12s achieves a 112% improvement in mAP over the baseline on COD10K-D and sets new state-of-the-art results for real-time camouflaged object detection across COD-D datasets.
Authors: Yi Gu, Zhaorui Wang, Jiahang Cao, Jiaxu Wang, Mingle Zhao, Dongjun Ye, Renjing Xu
Abstract: 3D Gaussian Splatting (3DGS) enables efficient training and fast novel view synthesis in static environments. To address challenges posed by transient objects, distractor-free 3DGS methods have emerged and shown promising results when dense image captures are available. However, their performance degrades significantly under sparse input conditions. This limitation primarily stems from the reliance on the color residual heuristics to guide the training, which becomes unreliable with limited observations. In this work, we propose a framework to enhance distractor-free 3DGS under sparse-view conditions by incorporating rich prior information. Specifically, we first adopt the geometry foundation model VGGT to estimate camera parameters and generate a dense set of initial 3D points. Then, we harness the attention maps from VGGT for efficient and accurate semantic entity matching. Additionally, we utilize Vision-Language Models (VLMs) to further identify and preserve the large static regions in the scene. We also demonstrate how these priors can be seamlessly integrated into existing distractor-free 3DGS methods. Extensive experiments confirm the effectiveness and robustness of our approach in mitigating transient distractors for sparse-view 3DGS training.
Authors: Qin Su, Tie Luo
Abstract: Vision Transformers (ViTs) achieve strong performance in visual recognition, yet their decision-making remains difficult to interpret. We propose BiCAM, a bidirectional class activation mapping method that captures both supportive (positive) and suppressive (negative) contributions to model predictions. Unlike prior CAM-based approaches that discard negative signals, BiCAM preserves signed attributions to produce more complete and contrastive explanations. BiCAM further introduces a Positive-to-Negative Ratio (PNR) that summarizes attribution balance and enables lightweight detection of adversarial examples without retraining. Across ImageNet, VOC, and COCO, BiCAM improves localization and faithfulness while remaining computationally efficient. It generalizes to multiple ViT variants, including DeiT and Swin. These results suggest the importance of modeling both supportive and suppressive evidence for interpreting transformer-based vision models.
Authors: Yuchen Zou, Xiao Hu, Dexing Zhong, Yuqing Tang
Abstract: Monocular re-localization plays a crucial role in enabling intelligent agents to achieve human-like perception. However, traditional methods rely on dense maps, which face scalability limitations and privacy risks. OpenStreetMap (OSM), as a lightweight map that protects privacy, offers semantic and geometric information with global scalability. Nonetheless, there are still challenges in using OSM for localization: the inherent cross-modal discrepancies between natural images and OSM, as well as the high computational cost of global map-based localization. In this paper, we propose a hierarchical search framework with semantic alignment for localization in OSM. First, the semantic awareness capability of DINO-ViT is utilised to deconstruct visual elements to establish semantic relationships with OSM. Second, a coarse-to-fine search paradigm is designed to replace global dense matching, enabling efficient progressive refinement. Extensive experiments demonstrate that our method significantly improves both localization accuracy and speed. When trained on a single dataset, the 3{\deg} orientation recall of our method even outperforms the 5{\deg} recall of state-of-the-art methods.
Authors: Jiaqi Han, Juntong Shi, Puheng Li, Haotian Ye, Qiushan Guo, Stefano Ermon
Abstract: Diffusion models have become the dominant tool for high-fidelity image and video generation, yet are critically bottlenecked by their inference speed due to the numerous iterative passes of Diffusion Transformers. To reduce the exhaustive compute, recent works resort to the feature caching and reusing scheme that skips network evaluations at selected diffusion steps by using cached features in previous steps. However, their preliminary design solely relies on local approximation, causing errors to grow rapidly with large skips and leading to degraded sample quality at high speedups. In this work, we propose spectral diffusion feature forecaster (Spectrum), a training-free approach that enables global, long-range feature reuse with tightly controlled error. In particular, we view the latent features of the denoiser as functions over time and approximate them with Chebyshev polynomials. Specifically, we fit the coefficient for each basis via ridge regression, which is then leveraged to forecast features at multiple future diffusion steps. We theoretically reveal that our approach admits more favorable long-horizon behavior and yields an error bound that does not compound with the step size. Extensive experiments on various state-of-the-art image and video diffusion models consistently verify the superiority of our approach. Notably, we achieve up to 4.79$\times$ speedup on FLUX.1 and 4.67$\times$ speedup on Wan2.1-14B, while maintaining much higher sample quality compared with the baselines.
Authors: Enhui Ma, Jiahuan Zhang, Guantian Zheng, Tao Tang, Shengbo Eben Li, Yuhang Lu, Xia Zhou, Xueyang Zhang, Yifei Zhan, Kun Zhan, Zhihui Hao, Xianpeng Lang, Kaicheng Yu
Abstract: Multimodal Large Language Models (MLLMs) are rapidly becoming the intelligence brain of end-to-end autonomous driving systems. A key challenge is to assess whether MLLMs can truly understand and follow complex real-world traffic rules. However, existing benchmarks mainly focus on single-rule scenarios like traffic sign recognition, neglecting the complexity of multi-rule concurrency and conflicts in real driving. Consequently, models perform well on simple tasks but often fail or violate rules in real world complex situations. To bridge this gap, we propose DriveCombo, a text and vision-based benchmark for compositional traffic rule reasoning. Inspired by human drivers' cognitive development, we propose a systematic Five-Level Cognitive Ladder that evaluates reasoning from single-rule understanding to multi-rule integration and conflict resolution, enabling quantitative assessment across cognitive stages. We further propose a Rule2Scene Agent that maps language-based traffic rules to dynamic driving scenes through rule crafting and scene generation, enabling scene-level traffic rule visual reasoning. Evaluations of 14 mainstream MLLMs reveal performance drops as task complexity grows, particularly during rule conflicts. After splitting the dataset and fine-tuning on the training set, we further observe substantial improvements in both traffic rule reasoning and downstream planning capabilities. These results highlight the effectiveness of DriveCombo in advancing compliant and intelligent autonomous driving systems.
Authors: Xiangyang He, Lin Wan
Abstract: Cloth-Changing Person Re-Identification (CC-ReID) aims to match the same individual across cameras under varying clothing conditions. Existing approaches often remove apparel and focus on the head region to reduce clothing bias. However, treating the head holistically without distinguishing between face and hair leads to over-reliance on volatile hairstyle cues, causing performance degradation under hairstyle changes. To address this issue, we propose the Mitigating Hairstyle Distraction and Structural Preservation (MSP) framework. Specifically, MSP introduces Hairstyle-Oriented Augmentation (HSOA), which generates intra-identity hairstyle diversity to reduce hairstyle dependence and enhance attention to stable facial and body cues. To prevent the loss of structural information, we design Cloth-Preserved Random Erasing (CPRE), which performs ratio-controlled erasing within clothing regions to suppress texture bias while retaining body shape and context. Furthermore, we employ Region-based Parsing Attention (RPA) to incorporate parsing-guided priors that highlight face and limb regions while suppressing hair features. Extensive experiments on multiple CC-ReID benchmarks demonstrate that MSP achieves state-of-the-art performance, providing a robust and practical solution for long-term person re-identification.
Authors: Rundong Wang, Wei Ba, Ying Zhou, Yingtai Li, Bowen Liu, Baizhi Wang, Yuhao Wang, Zhidong Yang, Kun Zhang, Rui Yan, S. Kevin Zhou
Abstract: Recent methods for pathology report generation from whole-slide image (WSI) are capable of producing slide-level diagnostic descriptions but fail to ground fine-grained statements in localized visual evidence. Furthermore, they lack control over which diagnostic details to include and how to verify them. Inspired by emerging agentic analysis paradigms and the diagnostic workflow of pathologists,who selectively examine multiple fields of view, we propose QCAgent, an agentic framework for quality-controllable WSI report generation. The core innovations of this framework are as follows: (i) it incorporates a customized critique mechanism guided by a user-defined checklist specifying required diagnostic details and constraints; (ii) it re-identifies informative regions in the WSI based on the critique feedback and text-patch semantic retrieval, a process that iteratively enriches and reconciles the report. Experiments demonstrate that by making report requirements explicitly prompt-defined, constraint-aware, and verifiable through evidence-grounded refinement, QCAgent enables controllable generation of clinically meaningful and high-coverage pathology reports from WSI.
Authors: Xianqi Wang, Hao Yang, Hangtian Wang, Junda Cheng, Gangwei Xu, Min Lin, Xin Yang
Abstract: Modern stereo matching methods have leveraged monocular depth foundation models to achieve superior zero-shot generalization performance. However, most existing methods primarily focus on extracting robust features for cost volume construction or disparity initialization. At the same time, the iterative refinement stage, which is also crucial for zero-shot generalization, remains underexplored. Some methods treat monocular depth priors as guidance for iteration, but conventional GRU-based architectures struggle to exploit them due to the limited representation capacity. In this paper, we propose Prompt Recurrent Unit (PRU), a novel iterative refinement module based on the decoder of monocular depth foundation models. By integrating monocular structure and stereo motion cues as prompts into the decoder, PRU enriches the latent representations of monocular depth foundation models with absolute stereo-scale information while preserving their inherent monocular depth priors. Experiments demonstrate that our PromptStereo achieves state-of-the-art zero-shot generalization performance across multiple datasets, while maintaining comparable or faster inference speed. Our findings highlight prompt-guided iterative refinement as a promising direction for zero-shot stereo matching.
Authors: Aryan Goyal, Shreshtha Singh, Ashish Mittal, Manoj Tadepalli, Piyush Kumar, Preetham Putha
Abstract: Early detection of lung cancer in chest radiographs (CXRs) is crucial for improving patient outcomes, yet nodule detection remains challenging due to their subtle appearance and variability in radiological characteristics like size, texture, and boundary. For robust analysis, this diversity must be well represented in training datasets for deep learning based Computer-Assisted Diagnosis (CAD) systems. However, assembling such datasets is costly and often impractical, motivating the need for realistic synthetic data generation. Existing methods lack fine-grained control over synthetic nodule generation, limiting their utility in addressing data scarcity. This paper proposes a novel diffusion-based framework with low-rank adaptation (LoRA) adapters for characteristic controlled nodule synthesis on CXRs. We begin by addressing size and shape control through nodule mask conditioned training of the base diffusion model. To achieve individual characteristic control, we train separate LoRA modules, each dedicated to a specific radiological feature. However, since nodules rarely exhibit isolated characteristics, effective multi-characteristic control requires a balanced integration of features. We address this by leveraging the dynamic composability of LoRAs and revisiting existing merging strategies. Building on this, we identify two key issues, overlapping attention regions and non-orthogonal parameter spaces. To overcome these limitations, we introduce a novel orthogonality loss term during LoRA composition training. Extensive experiments on both in-house and public datasets demonstrate improved downstream nodule detection. Radiologist evaluations confirm the fine-grained controllability of our generated nodules, and across multiple quantitative metrics, our method surpasses existing nodule generation approaches for CXRs.
Authors: Shao Shitong, Gu Yufei, Xie Zeke
Abstract: The recent advent of powerful video generation models, such as Hunyuan, WanX, Veo3, and Kling, has inaugurated a new era in the field. However, the practical deployment of these models is severely impeded by their substantial computational overhead, which stems from enormous parameter counts and the iterative, multi-step sampling process required during inference. Prior research on accelerating generative models has predominantly followed two distinct trajectories: reducing the number of sampling steps (e.g., LCM, DMD, and MagicDistillation) or compressing the model size for more efficient inference (e.g., ICMD). The potential of simultaneously compressing both to create a fast and lightweight model remains an unexplored avenue. In this paper, we propose FastLightGen, an algorithm that transforms large, computationally expensive models into fast, lightweight counterparts. The core idea is to construct an optimal teacher model, one engineered to maximize student performance, within a synergistic framework for distilling both model size and inference steps. Our extensive experiments on HunyuanVideo-ATI2V and WanX-TI2V reveal that a generator using 4-step sampling and 30\% parameter pruning achieves optimal visual quality under a constrained inference budget. Furthermore, FastLightGen consistently outperforms all competing methods, establishing a new state-of-the-art in efficient video generation.
Authors: Aryan Goyal, Ashish Mittal, Pranav Rao, Manoj Tadepalli, Preetham Putha
Abstract: Deep learning-based automated diagnosis of lung cancer has emerged as a crucial advancement that enables healthcare professionals to detect and initiate treatment earlier. However, these models require extensive training datasets with diverse case-specific properties. High-quality annotated data is particularly challenging to obtain, especially for cases with subtle pulmonary nodules that are difficult to detect even for experienced radiologists. This scarcity of well-labeled datasets can limit model performance and generalization across different patient populations. Digitally reconstructed radiographs (DRR) using CT-Scan to generate synthetic frontal chest X-rays with artificially inserted lung nodules offers one potential solution. However, this approach suffers from significant image quality degradation, particularly in the form of blurred anatomical features and loss of fine lung field structures. To overcome this, we introduce DiffusionXRay, a novel image restoration pipeline for Chest X-ray images that synergistically leverages denoising diffusion probabilistic models (DDPMs) and generative adversarial networks (GANs). DiffusionXRay incorporates a unique two-stage training process: First, we investigate two independent approaches, DDPM-LQ and GAN-based MUNIT-LQ, to generate low-quality CXRs, addressing the challenge of training data scarcity, posing this as a style transfer problem. Subsequently, we train a DDPM-based model on paired low-quality and high-quality images, enabling it to learn the nuances of X-ray image restoration. Our method demonstrates promising results in enhancing image clarity, contrast, and overall diagnostic value of chest X-rays while preserving subtle yet clinically significant artifacts, validated by both quantitative metrics and expert radiological assessment.
Authors: Gong Chen, Chaokun Zhang, Pengcheng Lv
Abstract: Cooperative perception lets agents share information to expand coverage and improve scene understanding. However, in real-world scenarios, diverse and unpredictable corruptions undermine its robustness and generalization. To address these challenges, we introduce CoopDiff, a diffusion-based cooperative perception framework that mitigates corruptions via a denoising mechanism. CoopDiff adopts a teacher-student paradigm: the Quality-Aware Teacher performs voxel-level early fusion with Quality of Interest weighting and semantic guidance, then produces clean supervision features via a diffusion denoiser. The Dual-Branch Diffusion Student first separates ego and cooperative streams in encoding to reconstruct the teacher's clean targets. And then, an Ego-Guided Cross-Attention mechanism facilitates balanced decoding under degradation by adaptively integrating ego and cooperative features. We evaluate CoopDiff on two constructed multi-degradation benchmarks, OPV2Vn and DAIR-V2Xn, each incorporating six corruption types, including environmental and sensor-level distortions. Benefiting from the inherent denoising properties of diffusion, CoopDiff consistently outperforms prior methods across all degradation types and lowers the relative corruption error. Furthermore, it offers a tunable balance between precision and inference efficiency.
Authors: Lirui Luo, Guoxi Zhang, Hongming Xu, Yaodong Yang, Cong Fang, Qing Li
Abstract: Reward design is of great importance for solving complex tasks with reinforcement learning. Recent studies have explored using image-text similarity produced by vision-language models (VLMs) to augment rewards of a task with visual feedback. A common practice linearly adds VLM scores to task or success rewards without explicit shaping, potentially altering the optimal policy. Moreover, such approaches, often relying on single static images, struggle with tasks whose desired behavior involves complex, dynamic motions spanning multiple visually different states. Furthermore, single viewpoints can occlude critical aspects of an agent's behavior. To address these issues, this paper presents Multi-View Video Reward Shaping (MVR), a framework that models the relevance of states regarding the target task using videos captured from multiple viewpoints. MVR leverages video-text similarity from a frozen pre-trained VLM to learn a state relevance function that mitigates the bias towards specific static poses inherent in image-based methods. Additionally, we introduce a state-dependent reward shaping formulation that integrates task-specific rewards and VLM-based guidance, automatically reducing the influence of VLM guidance once the desired motion pattern is achieved. We confirm the efficacy of the proposed framework with extensive experiments on challenging humanoid locomotion tasks from HumanoidBench and manipulation tasks from MetaWorld, verifying the design choices through ablation studies.
Authors: Haonan Jia, Shichao Dong, Xin Dong, Zenghui Sun, Jin Wang, Jinsong Lan, Xiaoyong Zhu, Bo Zheng, Kaifu Zhang
Abstract: Large Vision-Language Models (LVLMs) often omit or misrepresent critical visual content in generated image captions. Minimizing such information loss will force LVLMs to focus on image details to generate precise descriptions. However, measuring information loss during modality conversion is inherently challenging due to the modal gap between visual content and text output. In this paper, we argue that the quality of an image caption is positively correlated with the similarity between images retrieved via text search using that caption. Based on this insight, we further propose Cross-modal Identity Mapping (CIM), a reinforcement learning framework that enhances image captioning without requiring additional annotations. Specifically, the method quantitatively evaluates the information loss from two perspectives: Gallery Representation Consistency and Query-gallery Image Relevance. Supervised under these metrics, LVLM minimizes information loss and aims to achieve identity mapping from images to captions. The experimental results demonstrate the superior performance of our method in image captioning, even when compared with Supervised Fine-Tuning. Particularly, on the COCO-LN500 benchmark, CIM achieves a 20% improvement in relation reasoning on Qwen2.5-VL-7B.The code will be released when the paper is accepted.
Authors: Ruixi Wu, Shaobo Wang, Jiahuan Chen, Zhiyuan Liu, Yicun Yang, Zhaorun Chen, Zekai Li, Kaixin Li, Xinming Wang, Hongzhu Yi, Kai Wang, Linfeng Zhang
Abstract: Dataset distillation (DD) aims to compress large-scale datasets into compact synthetic counterparts for efficient model training. However, existing DD methods exhibit substantial performance degradation on long-tailed datasets. We identify two fundamental challenges: heuristic design choices for distribution discrepancy measure and uniform treatment of imbalanced classes. To address these limitations, we propose Class-Aware Spectral Distribution Matching (CSDM), which reformulates distribution alignment via the spectrum of a well-behaved kernel function. This technique maps the original samples into frequency space, resulting in the Spectral Distribution Distance (SDD). To mitigate class imbalance, we exploit the unified form of SDD to perform amplitude-phase decomposition, which adaptively prioritizes the realism in tail classes. On CIFAR-10-LT, with 10 images per class, CSDM achieves a 14.0% improvement over state-of-the-art DD methods, with only a 5.7% performance drop when the number of images in tail classes decreases from 500 to 25, demonstrating strong stability on long-tailed data.
Authors: Tianqi Shen, Huakao Lin, Ning An
Abstract: Siamese visual trackers have recently advanced through increasingly sophisticated fusion mechanisms built on convolutional or Transformer architectures. However, both struggle to deliver pixel-level interactions efficiently on resource-constrained hardware, leading to a persistent accuracy-efficiency imbalance. Motivated by this limitation, we redesign the Siamese neck with a simple yet effective Multilayer Perception (MLP)-based fusion module that enables pixel-level interaction with minimal structural overhead. Nevertheless, naively stacking MLP blocks introduces a new challenge: computational cost can scale quadratically with channel width. To overcome this, we construct a hierarchical search space of carefully designed MLP modules and introduce a customized relaxation strategy that enables differentiable neural architecture search (DNAS) to decouple channel-width optimization from other architectural choices. This targeted decoupling automatically balances channel width and depth, yielding a low-complexity architecture. The resulting tracker achieves state-of-the-art accuracy-efficiency trade-offs. It ranks among the top performers on four general-purpose and three aerial tracking benchmarks, while maintaining real-time performance on both resource-constrained Graphics Processing Units (GPUs) and Neural Processing Units (NPUs).
Authors: Gong Chen, Chaokun Zhang, Xinyan Zhao
Abstract: Collaborative perception is vital for autonomous driving yet remains constrained by tight communication budgets. Earlier work reduced bandwidth by compressing full feature maps with fixed-rate encoders, which adapts poorly to a changing environment, and it further evolved into spatial selection methods that improve efficiency by focusing on salient regions, but this object-centric approach often sacrifices global context, weakening holistic scene understanding. To overcome these limitations, we introduce \textit{WhisperNet}, a bandwidth-aware framework that proposes a novel, receiver-centric paradigm for global coordination across agents. Senders generate lightweight saliency metadata, while the receiver formulates a global request plan that dynamically budgets feature contributions across agents and features, retrieving only the most informative features. A collaborative feature routing module then aligns related messages before fusion to ensure structural consistency. Extensive experiments show that WhisperNet achieves state-of-the-art performance, improving AP@0.7 on OPV2V by 2.4\% with only 0.5\% of the communication cost. As a plug-and-play component, it boosts strong baselines with merely 5\% of full bandwidth while maintaining robustness under localization noise. These results demonstrate that globally-coordinated allocation across \textit{what} and \textit{where} to share is the key to achieving efficient collaborative perception.
Authors: Le Dong, Qinzhong Tan, Chunlei Li, Jingliang Hu, Yilei Shi, Weisheng Dong, Xiao Xiang Zhu, Lichao Mou
Abstract: Anomaly detection is a critical task in computer vision with profound implications for medical imaging, where identifying pathologies early can directly impact patient outcomes. While recent unsupervised anomaly detection approaches show promise, they require substantial normal training data and struggle to generalize across anatomical contexts. We introduce D$^2$4FAD, a novel dual distillation framework for few-shot anomaly detection that identifies anomalies in previously unseen tasks using only a small number of normal reference images. Our approach leverages a pre-trained encoder as a teacher network to extract multi-scale features from both support and query images, while a student decoder learns to distill knowledge from the teacher on query images and self-distill on support images. We further propose a learn-to-weight mechanism that dynamically assesses the reference value of each support image conditioned on the query, optimizing anomaly detection performance. To evaluate our method, we curate a comprehensive benchmark dataset comprising 13,084 images across four organs, four imaging modalities, and five disease categories. Extensive experiments demonstrate that D$^2$4FAD significantly outperforms existing approaches, establishing a new state-of-the-art in few-shot medical anomaly detection. Code is available at https://github.com/ttttqz/D24FAD.
Authors: Ruize Cui, Jialun Pei, Haiqiao Wang, Jun Zhou, Jeremy Yuen-Chun Teoh, Pheng-Ann Heng, Jing Qin
Abstract: In laparoscopic liver surgery, augmented reality technology enhances intraoperative anatomical guidance by overlaying 3D liver models from preoperative CT/MRI onto laparoscopic 2D views. However, existing registration methods lack explicit modeling of reliable 2D-3D geometric correspondences supported by latent evidence, leading to limited interpretability and potentially unstable alignment in clinical scenarios. In this work, we introduce Land-Reg, a correspondence-driven deformable registration framework that explicitly learns latent-grounded 2D-3D landmark correspondences as an interpretable intermediate representation to bridge cross-modal alignment. For rigid registration, Land-Reg embraces a Cross-modal Latent Alignment module to map multi-modal features into a unified latent space. Further, an Uncertainty-enhanced Overlap Landmark Detector with similarity matching is proposed to robustly estimate explicit 2D-3D landmark correspondences. For non-rigid registration, we design a novel shape-constrained supervision strategy that anchors shape deformation to matched landmarks through reprojection consistency and incorporates local-isometric regularization to alleviate inherent 2D-3D depth ambiguity, while a rendered-mask alignment enforces global shape consistency. Experimental results on the P2ILF dataset demonstrate the superiority of our method on both rigid pose estimation and non-rigid deformation. Our code will be available at https://github.com/cuiruize/Land-Reg.
Authors: Guanglu Dong, Chunlei Li, Chao Ren, Jingliang Hu, Yilei Shi, Xiao Xiang Zhu, Lichao Mou
Abstract: Recently, significant breakthroughs have been made in all-in-one image restoration (AiOIR), which can handle multiple restoration tasks with a single model. However, existing methods typically focus on a specific image domain, such as natural scene, medical imaging, or remote sensing. In this work, we aim to extend AiOIR to multiple domains and propose the first multi-domain all-in-one image restoration method, DATPRL-IR, based on our proposed Domain-Aware Task Prompt Representation Learning. Specifically, we first construct a task prompt pool containing multiple task prompts, in which task-related knowledge is implicitly encoded. For each input image, the model adaptively selects the most relevant task prompts and composes them into an instance-level task representation via a prompt composition mechanism (PCM). Furthermore, to endow the model with domain awareness, we introduce another domain prompt pool and distill domain priors from multimodal large language models into the domain prompts. PCM is utilized to combine the adaptively selected domain prompts into a domain representation for each input image. Finally, the two representations are fused to form a domain-aware task prompt representation which can make full use of both specific and shared knowledge across tasks and domains to guide the subsequent restoration process. Extensive experiments demonstrate that our DATPRL-IR significantly outperforms existing SOTA image restoration methods, while exhibiting strong generalization capabilities. Code is available at https://github.com/GuangluDong0728/DATPRL-IR.
Authors: Tsung-Ming Tai, Sofia Casarin, Andrea Pilzer, Werner Nutt, Oswald Lanz
Abstract: Anticipating future actions in videos is challenging, as the observed frames provide only evidence of past activities, requiring the inference of latent intentions to predict upcoming actions. Existing transformer-based approaches, which rely on dot-product attention over pixel representations, often lack the high-level semantics necessary to model video sequences for effective action anticipation. As a result, these methods tend to overfit to explicit visual cues present in the past frames, limiting their ability to capture underlying intentions and degrading generalization to unseen samples. To address this, we propose Action-Guided Attention (AGA), an attention mechanism that explicitly leverages predicted action sequences as queries and keys to guide sequence modeling. Our approach fosters the attention module to emphasize relevant moments from the past based on the upcoming activity and combine this information with the current frame embedding via a dedicated gating function. The design of AGA enables post-training analysis of the knowledge discovered from the training set. Experiments on the widely adopted EPIC-Kitchens-100 benchmark demonstrate that AGA generalizes well from validation to unseen test sets. Post-training analysis can further examine the action dependencies captured by the model and the counterfactual evidence it has internalized, offering transparent and interpretable insights into its anticipative predictions.
Authors: Alexandru Manole, Laura Diosan
Abstract: Most information in our world is organized hierarchically; however, many Deep Learning approaches do not leverage this semantically rich structure. Research suggests that human learning benefits from exploiting the hierarchical structure of information, and intelligent models could similarly take advantage of this through multi-task learning. In this work, we analyze the advantages and limitations of multi-task learning in a hierarchical multi-label classification problem: car make and model classification. Considering both parallel and cascaded multi-task architectures, we evaluate their impact on different Deep Learning classifiers (CNNs, Transformers) while varying key factors such as dropout rate and loss weighting to gain deeper insight into the effectiveness of this approach. The tests are conducted on two established benchmarks: StanfordCars and CompCars. We observe the effectiveness of the multi-task paradigm on both datasets, improving the performance of the investigated CNN in almost all scenarios. Furthermore, the approach yields significant improvements on the CompCars dataset for both types of models.
Authors: Rong Fu, Yiqing Lyu, Chunlei Meng, Muge Qi, Yabin Jin, Qi Zhao, Li Bao, Juntao Gao, Fuqian Shi, Nilanjan Dey, Wei Luo, Simon Fong
Abstract: Automatic generation of radiology reports seeks to reduce clinician workload while improving documentation consistency. Existing methods that adopt encoder-decoder or retrieval-augmented pipelines achieve progress in fluency but remain vulnerable to visual-linguistic biases, factual inconsistency, and lack of explicit multi-hop clinical reasoning. We present NeuroSymb-MRG, a unified framework that integrates NeuroSymbolic abductive reasoning with active uncertainty minimization to produce structured, clinically grounded reports. The system maps image features to probabilistic clinical concepts, composes differentiable logic-based reasoning chains, decodes those chains into templated clauses, and refines the textual output via retrieval and constrained language-model editing. An active sampling loop driven by rule-level uncertainty and diversity guides clinician-in-the-loop adjudication and promptbook refinement. Experiments on standard benchmarks demonstrate consistent improvements in factual consistency and standard language metrics compared to representative baselines.
Authors: Keli Liu, Zhendong Wang, Wengang Zhou, Houqiang Li
Abstract: Visual AutoRegressive (VAR) models based on next-scale prediction enable efficient hierarchical generation, yet the inference cost grows quadratically at high resolutions. We observe that the computationally intensive later scales predominantly refine high-frequency textures and exhibit substantial spatial redundancy, in contrast to earlier scales that determine the global structural layout. Existing pruning methods primarily focus on high-frequency detection for token selection, often overlooking structural coherence and consequently degrading global semantics. To address this limitation, we propose StepVAR, a training-free token pruning framework that accelerates VAR inference by jointly considering structural and textural importance. Specifically, we employ a lightweight high-pass filter to capture local texture details, while leveraging Principal Component Analysis (PCA) to preserve global structural information. This dual-criterion design enables the model to retain tokens critical for both fine-grained fidelity and overall composition. To maintain valid next-scale prediction under sparse tokens, we further introduce a nearest neighbor feature propagation strategy to reconstruct dense feature maps from pruned representations. Extensive experiments on state-of-the-art text-to-image and text-to-video VAR models demonstrate that StepVAR achieves substantial inference speedups while maintaining generation quality. Quantitative and qualitative evaluations consistently show that our method outperforms existing acceleration approaches, validating its effectiveness and general applicability across diverse VAR architectures.
Authors: Yuxuan Li, Yuming Chen, Yunheng Li, Ming-Ming Cheng, Xiang Li, Jian Yang
Abstract: Heterogeneous multi-modal remote sensing object detection aims to accurately detect objects from diverse sensors (e.g., RGB, SAR, Infrared). Existing approaches largely adopt a late alignment paradigm, in which modality alignment and task-specific optimization are entangled during downstream fine-tuning. This tight coupling complicates optimization and often results in unstable training and suboptimal generalization. To address these limitations, we propose BabelRS, a unified language-pivoted pretraining framework that explicitly decouples modality alignment from downstream task learning. BabelRS comprises two key components: Concept-Shared Instruction Aligning (CSIA) and Layerwise Visual-Semantic Annealing (LVSA). CSIA aligns each sensor modality to a shared set of linguistic concepts, using language as a semantic pivot to bridge heterogeneous visual representations. To further mitigate the granularity mismatch between high-level language representations and dense detection objectives, LVSA progressively aggregates multi-scale visual features to provide fine-grained semantic guidance. Extensive experiments demonstrate that BabelRS stabilizes training and consistently outperforms state-of-the-art methods without bells and whistles. Code: https://github.com/zcablii/SM3Det.
Authors: Minseok Seo, Wonjun Lee, Jaehyuk Jang, Changick Kim
Abstract: Zero-shot depth completion has gained attention for its ability to generalize across environments without sensor-specific datasets or retraining. However, most existing approaches rely on diffusion-based test-time optimization, which is computationally expensive due to iterative denoising. Recent visual-prompt-based methods reduce training cost but still require repeated forward--backward passes through the full frozen network to optimize input-level prompts, resulting in slow inference. In this work, we show that adapting only the decoder is sufficient for effective test-time optimization, as depth foundation models concentrate depth-relevant information within a low-dimensional decoder subspace. Based on this insight, we propose a lightweight test-time adaptation method that updates only this low-dimensional subspace using sparse depth supervision. Our approach achieves state-of-the-art performance, establishing a new Pareto frontier between accuracy and efficiency for test-time adaptation. Extensive experiments on five indoor and outdoor datasets demonstrate consistent improvements over prior methods, highlighting the practicality of fast zero-shot depth completion.
Authors: Bosen Lin, Feng Gao, Yanwei Yu, Junyu Dong, Qian Du
Abstract: In real underwater environments, downstream image recognition tasks such as semantic segmentation and object detection often face challenges posed by problems like blurring and color inconsistencies. Underwater image enhancement (UIE) has emerged as a promising preprocessing approach, aiming to improve the recognizability of targets in underwater images. However, most existing UIE methods mainly focus on enhancing images for human visual perception, frequently failing to reconstruct high-frequency details that are critical for task-specific recognition. To address this issue, we propose a Downstream Task-Inspired Underwater Image Enhancement (DTI-UIE) framework, which leverages human visual perception model to enhance images effectively for underwater vision tasks. Specifically, we design an efficient two-branch network with task-aware attention module for feature mixing. The network benefits from a multi-stage training framework and a task-driven perceptual loss. Additionally, inspired by human perception, we automatically construct a Task-Inspired UIE Dataset (TI-UIED) using various task-specific networks. Experimental results demonstrate that DTI-UIE significantly improves task performance by generating preprocessed images that are beneficial for downstream tasks such as semantic segmentation, object detection, and instance segmentation. The codes are publicly available at https://github.com/oucailab/DTIUIE.
Authors: Dragos Costea, Alina Marcu, Cristina Lazar, Marius Leordeanu
Abstract: We study the ongoing debate regarding the statistical fidelity of AI-generated data compared to human-generated data in the context of non-verbal communication using full body motion. Concretely, we ask if contemporary generative models move beyond surface mimicry to participate in the silent, but expressive dialogue of body language. We tackle this question by introducing the first framework that generates a natural non-verbal interaction between Human and AI in real-time from 2D body keypoints. Our experiments utilize four lightweight architectures which run at up to 100 FPS on an NVIDIA Orin Nano, effectively closing the perception-action loop needed for natural Human-AI interaction. We trained on 437 human video clips and demonstrated that pretraining on synthetically-generated sequences reduces motion errors significantly, without sacrificing speed. Yet, a measurable reality gap persists. When the best model is evaluated on keypoints extracted from cutting-edge text-to-video systems, such as SORA and VEO, we observe that performance drops on SORA-generated clips. However, it degrades far less on VEO, suggesting that temporal coherence, not image fidelity, drives real-world performance. Our results demonstrate that statistically distinguishable differences persist between Human and AI motion.
Authors: Ruoyang Su, Xi-Le Zhao, Sheng Liu, Wei-Hao Wu, Yisi Luo, Michael K. Ng
Abstract: Recently, continuous tensor functions have attracted increasing attention, because they can unifiedly represent data both on mesh grids and beyond mesh grids. However, since mode-$n$ product is essentially discrete and linear, the potential of current continuous tensor function representations is still locked. To break this bottleneck, we suggest neural operator-grounded mode-$n$ operators as a continuous and nonlinear alternative of discrete and linear mode-$n$ product. Instead of mapping the discrete core tensor to the discrete target tensor, proposed mode-$n$ operator directly maps the continuous core tensor function to the continuous target tensor function, which provides a genuine continuous representation of real-world data and can ameliorate discretization artifacts. Empowering with continuous and nonlinear mode-$n$ operators, we propose a neural operator-grounded continuous tensor function representation (abbreviated as NO-CTR), which can more faithfully represent complex real-world data compared with classic discrete tensor representations and continuous tensor function representations. Theoretically, we also prove that any continuous tensor function can be approximated by NO-CTR. To examine the capability of NO-CTR, we suggest an NO-CTR-based multi-dimensional data completion model. Extensive experiments across various data on regular mesh grids (multi-spectral images and color videos), on mesh girds with different resolutions (Sentinel-2 images) and beyond mesh grids (point clouds) demonstrate the superiority of NO-CTR.
Authors: Levente Hajder
Abstract: Affine transformations have been recently used for stereo vision. They can be exploited in various computer vision application, e.g., when estimating surface normals, homographies, fundamental and essential matrices. Even full 3D reconstruction can be obtained by using affine correspondences. First, this paper overviews the fundamental statements for affine transformations and epipolar geometry. Then it is investigated how the transformation accuracy influences the quality of the 3D reconstruction. Besides, we propose novel techniques for estimating the local affine transformation from corresponding image directions; moreover, the fundamental matrix, related to the processed image pair, can also be exploited. Both synthetic and real quantitative evaluations are implemented based on the accuracy of the reconstructed surface normals. For the latter one, a special object, containing three perpendicular planes with chessboard patterns, is constructed. The quantitative evaluations are based on the accuracy of the reconstructed surface normals and it is concluded that the estimation accuracy is around a few degrees for realistic test cases. Special stereo poses and plane orientations are also evaluated in detail.
Authors: Kuangyi Chen, Jun Zhang, Yuxi Hu, Yi Zhou, Friedrich Fraundorfer
Abstract: Event cameras offer high-temporal-resolution sensing that remains reliable under high-speed motion and challenging lighting, making them promising for localization from LiDAR point clouds in GPS-denied and visually degraded environments. However, aligning sparse, asynchronous events with dense LiDAR maps is fundamentally ill-posed, as direct correspondence estimation suffers from modality gaps. We propose LEAR, a dual-task learning framework that jointly estimates edge structures and dense event-depth flow fields to bridge the sensing-modality divide. Instead of treating edges as a post-hoc aid, LEAR couples them with flow estimation through a cross-modal fusion mechanism that injects modality-invariant geometric cues into the motion representation, and an iterative refinement strategy that enforces mutual consistency between the two tasks over multiple update steps. This synergy produces edge-aware, depth-aligned flow fields that enable more robust and accurate pose recovery via Perspective-n-Point (PnP) solvers. On several popular and challenging datasets, LEAR achieves superior performance over the best prior method. The source code, trained models, and demo videos are made publicly available online.
Authors: Hao Wu, Haoran Lou, Xinyue Li, Zuodong Zhong, Zhaojun Sun, Phellon Chen, Xuanhe Zhou, Kai Zuo, Yibo Chen, Xu Tang, Yao Hu, Boxiang Zhou, Jian Wu, Yongji Wu, Wenxin Yu, Yingmiao Liu, Yuhao Huang, Manjie Xu, Gang Liu, Yidong Ma, Zhichao Sun, Changhao Qiao
Abstract: We present FireRed-OCR, a systematic framework to specialize general VLMs into high-performance OCR models. Large Vision-Language Models (VLMs) have demonstrated impressive general capabilities but frequently suffer from ``structural hallucination'' when processing complex documents, limiting their utility in industrial OCR applications. In this paper, we introduce FireRed-OCR, a novel framework designed to transform general-purpose VLMs (based on Qwen3-VL) into pixel-precise structural document parsing experts. To address the scarcity of high-quality structured data, we construct a ``Geometry + Semantics'' Data Factory. Unlike traditional random sampling, our pipeline leverages geometric feature clustering and multi-dimensional tagging to synthesize and curate a highly balanced dataset, effectively handling long-tail layouts and rare document types. Furthermore, we propose a Three-Stage Progressive Training strategy that guides the model from pixel-level perception to logical structure generation. This curriculum includes: (1) Multi-task Pre-alignment to ground the model's understanding of document structure; (2) Specialized SFT for standardizing full-image Markdown output; and (3) Format-Constrained Group Relative Policy Optimization (GRPO), which utilizes reinforcement learning to enforce strict syntactic validity and structural integrity (e.g., table closure, formula syntax). Extensive evaluations on OmniDocBench v1.5 demonstrate that FireRed-OCR achieves state-of-the-art performance with an overall score of 92.94\%, significantly outperforming strong baselines such as DeepSeek-OCR 2 and OCRVerse across text, formula, table, and reading order metrics. We open-source our code and model weights to facilitate the ``General VLM to Specialized Structural Expert'' paradigm.
Authors: Yutong Yang, Katarina Popovi\'c, Julian Wiederer, Markus Braun, Vasileios Belagiannis, Bin Yang
Abstract: Detection Transformer (DETR) and its variants show strong performance on object detection, a key task for autonomous systems. However, a critical limitation of these models is that their confidence scores only reflect semantic uncertainty, failing to capture the equally important spatial uncertainty. This results in an incomplete assessment of the detection reliability. On the other hand, Deep Ensembles can tackle this by providing high-quality spatial uncertainty estimates. However, their immense memory consumption makes them impractical for real-world applications. A cheaper alternative, Monte Carlo (MC) Dropout, suffers from high latency due to the need of multiple forward passes during inference to estimate uncertainty. To address these limitations, we introduce GroupEnsemble, an efficient and effective uncertainty estimation method for DETR-like models. GroupEnsemble simultaneously predicts multiple individual detection sets by feeding additional diverse groups of object queries to the transformer decoder during inference. Each query group is transformed by the shared decoder in isolation and predicts a complete detection set for the same input. An attention mask is applied to the decoder to prevent inter-group query interactions, ensuring each group detects independently to achieve reliable ensemble-based uncertainty estimation. By leveraging the decoder's inherent parallelism, GroupEnsemble efficiently estimates uncertainty in a single forward pass without sequential repetition. We validated our method under autonomous driving scenes and common daily scenes using the Cityscapes and COCO datasets, respectively. The results show that a hybrid approach combining MC-Dropout and GroupEnsemble outperforms Deep Ensembles on several metrics at a fraction of the cost. The code is available at https://github.com/yutongy98/GroupEnsemble.
Authors: Alexander Prutsch, David Schinagl, Horst Possegger
Abstract: Future trajectories of neighboring traffic agents have a significant influence on the path planning and decision-making of autonomous vehicles. While trajectory forecasting is a well-studied field, research mainly focuses on snapshot-based prediction, where each scenario is treated independently of its global temporal context. However, real-world autonomous driving systems need to operate in a continuous setting, requiring real-time processing of data streams with low latency and consistent predictions over successive timesteps. We leverage this continuous setting to propose a lightweight yet highly accurate streaming-based trajectory forecasting approach. We integrate valuable information from previous predictions with a novel endpoint-aware modeling scheme. Our temporal context propagation uses the trajectory endpoints of the previous forecasts as anchors to extract targeted scenario context encodings. Our approach efficiently guides its scene encoder to extract highly relevant context information without needing refinement iterations or segment-wise decoding. Our experiments highlight that our approach effectively relays information across consecutive timesteps. Unlike methods using multi-stage refinement processing, our approach significantly reduces inference latency, making it well-suited for real-world deployment. We achieve state-of-the-art streaming trajectory prediction results on the Argoverse~2 multi-agent and single-agent benchmarks, while requiring substantially fewer resources.
Authors: Yiheng Li, Zichang Tan, Guoqing Xu, Yijun Ye, Yang Yang, Zhen Lei
Abstract: With the rapid development of generative AI in medical imaging, synthetic Computed Tomography (CT) images have demonstrated great potential in applications such as data augmentation and clinical diagnosis, but they also introduce serious security risks. Despite the increasing security concerns, existing studies on CT forgery detection are still limited and fail to adequately address real-world challenges. These limitations are mainly reflected in two aspects: the absence of datasets that can effectively evaluate model generalization to reflect the real-world application requirements, and the reliance on detection methods designed for natural images that are insensitive to CT-specific forgery artifacts. In this view, we propose CTForensics, a comprehensive dataset designed to systematically evaluate the generalization capability of CT forgery detection methods, which includes ten diverse CT generative methods. Moreover, we introduce the Enhanced Spatial-Frequency CT Forgery Detector (ESF-CTFD), an efficient CNN-based neural network that captures forgery cues across the wavelet, spatial, and frequency domains. First, it transforms the input CT image into three scales and extracts features at each scale via the Wavelet-Enhanced Central Stem. Then, starting from the largest-scale features, the Spatial Process Block gradually performs feature fusion with the smaller-scale ones. Finally, the Frequency Process Block learns frequency-domain information for predicting the final results. Experiments demonstrate that ESF-CTFD consistently outperforms existing methods and exhibits superior generalization across different CT generative models.
Authors: Muyu Liu, Xuanyu Tian, Chenhe Du, Qing Wu, Hongjiang Wei, Yuyao Zhang
Abstract: Recovering radiometric fidelity from unknown dynamic range compression (UDRC), such as low-light enhancement and HDR reconstruction, is a challenging blind inverse problem, due to the unknown forward model and irreversible information loss introduced by compression. To address this challenge, we first identify monotonicity as the fundamental physical invariant shared across UDRC tasks. Leveraging this insight, we introduce the \textbf{cascaded monotonic Bernstein} (CaMB) operator to parameterize the unknown forward model. CaMB enforces monotonicity as a hard architectural inductive bias, constraining optimization to physically consistent mappings and enabling robust and stable operator estimation. We further integrate CaMB with a plug-and-play diffusion framework, proposing \textbf{CaMB-Diff}. Within this framework, the diffusion model serves as a powerful geometric prior for structural and semantic recovery, while CaMB explicitly models and corrects radiometric distortions through a physically grounded forward operator. Extensive experiments on a variety of zero-shot UDRC tasks, including low-light enhancement, low-field MRI enhancement, and HDR reconstruction, demonstrate that CaMB-Diff significantly outperforms state-of-the-art zero-shot baselines in terms of both signal fidelity and physical consistency. Moreover, we empirically validate the effectiveness of the proposed CaMB parameterization in accurately modeling the unknown forward operator.
Authors: Zijin Yin, Tiankai Hang, Yiji Cheng, Shiyi Zhang, Runze He, Yu Xu, Chunyu Wang, Bing Li, Zheng Chang, Kongming Liang, Qinglin Lu, Zhanyu Ma
Abstract: Existing image editing methods struggle to perceive where to edit, especially under complex scenes and nuanced spatial instructions. To address this issue, we propose Generative Visual Chain-of-Thought (GVCoT), a unified framework that performs native visual reasoning by first generating spatial cues to localize the target region and then executing the edit. Unlike prior text-only CoT or tool-dependent visual CoT paradigms, GVCoT jointly optimizes visual tokens generated during the reasoning and editing phases in an end-to-end manner. This way fosters the emergence of innate spatial reasoning ability and enables more effective utilization of visual-domain cues. The main challenge of training GCVoT lies in the scarcity of large-scale editing data with precise edit region annotations; to this end, we construct GVCoT-Edit-Instruct, a dataset of 1.8M high-quality samples spanning 19 tasks. We adopt a progressive training strategy: supervised fine-tuning to build foundational localization ability in reasoning trace before final editing, followed by reinforcement learning to further improve reasoning and editing quality. Finally, we introduce SREdit-Bench, a new benchmark designed to comprehensively stress-test models under sophisticated scenes and fine-grained referring expressions. Experiments demonstrate that GVCoT consistently outperforms state-of-the-art models on SREdit-Bench and ImgEdit. We hope our GVCoT will inspire future research toward interpretable and precise image editing.
Authors: Muyu Liu, Chenhe Du, Xuanyu Tian, Qing Wu, Xiao Wang, Haonan Zhang, Hongjiang Wei, Yuyao Zhang
Abstract: Low-field (LF) magnetic resonance imaging (MRI) democratizes access to diagnostic imaging but is fundamentally limited by low signal-to-noise ratio and significant tissue contrast distortion due to field-dependent relaxation dynamics. Reconstructing high-field (HF) quality images from LF data is a blind inverse problem, severely challenged by the scarcity of paired training data and the unknown, non-linear contrast transformation operator. Existing zero-shot methods, which assume simplified linear degradation, often fail to recover authentic tissue contrast. In this paper, we propose DACT(Diffusion-Based Adaptive Contrast Transport), a novel zero-shot framework that restores HF-quality images without paired supervision. DACT synergizes a pre-trained HF diffusion prior to ensure anatomical fidelity with a physically-informed adaptive forward model. Specifically, we introduce a differentiable Sinkhorn optimal transport module that explicitly models and corrects the intensity distribution shift between LF and HF domains during the reverse diffusion process. This allows the framework to dynamically learn the intractable contrast mapping while preserving topological consistency. Extensive experiments on simulated and real clinical LF datasets demonstrate that DACT achieves state-of-the-art performance, yielding reconstructions with superior structural detail and correct tissue contrast.
Authors: Yuechen Luo, Fang Li, Shaoqing Xu, Yang Ji, Zehan Zhang, Bing Wang, Yuannan Shen, Jianwei Cui, Long Chen, Guang Chen, Hangjun Ye, Zhi-Xin Yang, Fuxi Wen
Abstract: While Vision-Language-Action (VLA) models have revolutionized autonomous driving by unifying perception and planning, their reliance on explicit textual Chain-of-Thought (CoT) leads to semantic-perceptual decoupling and perceptual-symbolic conflicts. Recent shifts toward latent reasoning attempt to bypass these bottlenecks by thinking in continuous hidden space. However, without explicit intermediate constraints, standard latent CoT often operates as a physics-agnostic representation. To address this, we propose the Latent Spatio-Temporal VLA (LaST-VLA), a framework shifting the reasoning paradigm from discrete symbolic processing into a physically grounded Latent Spatio-Temporal CoT. By implementing a dual-feature alignment mechanism, we distill geometric constraints from 3D foundation models and dynamic foresight from world models directly into the latent space. Coupled with a progressive SFT training strategy that transitions from feature alignment to trajectory generation, and refined via Reinforcement Learning with Group Relative Policy Optimization (GRPO) to ensure safety and rule compliance. \method~setting a new record on NAVSIM v1 (91.3 PDMS) and NAVSIM v2 (87.1 EPDMS), while excelling in spatial-temporal reasoning on SURDS and NuDynamics benchmarks.
Authors: Haitian Wang, Xinyu Wang, Muhammad Ibrahim, Dustin Severtson, Ajmal Mian
Abstract: Accurate weed mapping in cereal fields requires pixel-level segmentation from UAV imagery that remains reliable across fields, seasons, and illumination. Existing multispectral pipelines often depend on thresholded vegetation indices, which are brittle under radiometric drift and mixed crop--weed pixels, or on single-stream CNN and Transformer backbones that ingest stacked bands and indices, where radiance cues and normalized index cues interfere and reduce sensitivity to small weed clusters embedded in crop canopies. We propose VISA (Vegetation-Index and Spectral Attention), a two-stream segmentation network that decouples these cues and fuses them at native resolution. The radiance stream learns from calibrated five-band reflectance using residual spectral-spatial attention to preserve fine textures and row boundaries that are attenuated by ratio indices. The index stream operates on vegetation-index maps with windowed self-attention to model local structure efficiently, state-space layers to propagate field-scale context without quadratic attention cost, and Slot Attention to form stable region descriptors that improve discrimination of sparse weeds under canopy mixing. To support supervised training and deployment-oriented evaluation, we introduce BAWSeg, a four-year UAV multispectral dataset collected over commercial barley paddocks in Western Australia, providing radiometrically calibrated blue, green, red, red edge, and near-infrared orthomosaics, derived vegetation indices, and dense crop, weed, and other labels with leakage-free block splits. On BAWSeg, VISA achieves 75.6% mIoU and 63.5% weed IoU with 22.8M parameters, outperforming a multispectral SegFormer-B1 baseline by 1.2 mIoU and 1.9 weed IoU. Under cross-plot and cross-year protocols, VISA maintains 71.2% and 69.2% mIoU, respectively. The BAWSeg data, VISA code, and trained models will be released upon publication.
Authors: Dinh Nam Pham, Leonard Prokisch, Bennet Meyer, Jonas Thumbs
Abstract: Smartphone clip-on microscopes turn everyday devices into low-cost, portable imaging systems that can even reveal fungal structures at the microscopic level, enabling mold inspection beyond unaided visual checks. In this paper, we introduce MobileMold, an open smartphone-based microscopy dataset for food mold detection and food classification. MobileMold contains 4,941 handheld microscopy images spanning 11 food types, 4 smartphones, 3 microscopes, and diverse real-world conditions. Beyond the dataset release, we establish baselines for (i) mold detection and (ii) food-type classification, including a multi-task setting that predicts both attributes. Across multiple pretrained deep learning architectures and augmentation strategies, we obtain near-ceiling performance (accuracy = 0.9954, F1 = 0.9954, MCC = 0.9907), validating the utility of our dataset for detecting food spoilage. To increase transparency, we complement our evaluation with saliency-based visual explanations highlighting mold regions associated with the model's predictions. MobileMold aims to contribute to research on accessible food-safety sensing, mobile imaging, and exploring the potential of smartphones enhanced with attachments.
Authors: Yuting Wan, Liguo Sun, Jiuwu Hao, Zao Zhang, Pin LV
Abstract: Detecting water-surface targets for Unmanned Surface Vehicles (USVs) is challenging due to wave clutter, specular reflections, and weak appearance cues in long-range observations. Although 4D millimeter-wave radar complements cameras under degraded illumination, maritime radar point clouds are sparse and intermittent, with reflectivity attributes exhibiting heavy-tailed variations under scattering and multipath, making conventional fusion designs struggle to exploit radar cues effectively. We propose PhysFusion, a physics-informed radar-image detection framework for water-surface perception. The framework integrates: (1) a Physics-Informed Radar Encoder (PIR Encoder) with an RCS Mapper and Quality Gate, transforming per-point radar attributes into compact scattering priors and predicting point-wise reliability for robust feature learning under clutter; (2) a Radar-guided Interactive Fusion Module (RIFM) performing query-level radar-image fusion between semantically enriched radar features and multi-scale visual features, with the radar branch modeled by a dual-stream backbone including a point-based local stream and a transformer-based global stream using Scattering-Aware Self-Attention (SASA); and (3) a Temporal Query Aggregation module (TQA) aggregating frame-wise fused queries over a short temporal window for temporally consistent representations. Experiments on WaterScenes and FLOW demonstrate that PhysFusion achieves 59.7% mAP50:95 and 90.3% mAP50 on WaterScenes (T=5 radar history) using 5.6M parameters and 12.5G FLOPs, and reaches 94.8% mAP50 and 46.2% mAP50:95 on FLOW under radar+camera setting. Ablation studies quantify the contributions of PIR Encoder, SASA-based global reasoning, and RIFM.
Authors: Yand Wang, Chen Zhang, Lanyun Zhu, Yixin Chen, Qunbo Wang, Yutong Bai, Jurgen Germann, Yinghong Wen, Shuai Shao
Abstract: Preoperative improvement rate prediction for Parkinson's disease surgery is clinically important yet difficult because imaging signals are subtle and patients are heterogeneous. We address this setting, where only information available before surgery is used, and the goal is to predict patient-specific postoperative motor benefit. We present PreSight, a presurgical outcome model that fuses clinical priors with preoperative MRI and deformation-based morphometry (DBM) and adapts regional importance through a patient-specific weighting module. The model produces end-to-end, calibrated, decision-ready predictions with patient-level explanations. We evaluate PreSight on a real-world two-center cohort of 400 subjects with multimodal presurgical inputs and postoperative improvement labels. PreSight outperforms strong clinical, imaging-only, and multimodal baselines. It attains 88.89% accuracy on internal validation and 85.29% on an external-center test for responder classification and shows better probability calibration and higher decision-curve net benefit. Ablations and analyses confirm the contribution of DBM and the patient-specific weighting module and indicate that the model emphasizes disease-relevant regions in a patient-specific manner. These results demonstrate that integrating clinical prior knowledge with region-adaptive morphometry enables reliable presurgical decision support in routine practice.
Authors: Luu Le, Hoang-Loc Cao, Ha-Hieu Pham, Thanh-Huy Nguyen, Ulas Bagci
Abstract: White blood cell (WBC) classification is fundamental for hematology applications such as infection assessment, leukemia screening, and treatment monitoring. However, real-world WBC datasets present substantial appearance variations caused by staining and scanning conditions, as well as severe class imbalance in which common cell types dominate while rare but clinically important categories are underrepresented. To address these challenges, we propose a stain-normalized, decoupled training framework that first learns transferable representations using instance-balanced sampling, and then rebalances the classifier with class-aware sampling and a hybrid loss combining effective-number weighting and focal modulation. In inference stage, we further enhance robustness by ensembling various trained backbones with test-time augmentation. Our approach achieved the top rank on the leaderboard of the WBCBench 2026: Robust White Blood Cell Classification Challenge at ISBI 2026.
Authors: Yuchen Zhang, Yaxiong Wang, Kecheng Han, Yujiao Wu, Lianwei Wu, Li Zhu, Zhedong Zheng
Abstract: Recent advances in generative AI have significantly enhanced the realism of multimodal media manipulation, thereby posing substantial challenges to manipulation detection. Existing manipulation detection and grounding approaches predominantly focus on manipulation type classification under result-oriented supervision, which not only lacks interpretability but also tends to overfit superficial artifacts. In this paper, we argue that generalizable detection requires incorporating explicit forensic reasoning, rather than merely classifying a limited set of manipulation types, which fails to generalize to unseen manipulation patterns. To this end, we propose REFORM, a reasoning-driven framework that shifts learning from outcome fitting to process modeling. REFORM adopts a three-stage curriculum that first induces forensic rationales, then aligns reasoning with final judgments, and finally refines logical consistency via reinforcement learning. To support this paradigm, we introduce ROM, a large-scale dataset with rich reasoning annotations. Extensive experiments show that REFORM establishes new state-of-the-art performance with superior generalization, achieving 81.52% ACC on ROM, 76.65% ACC on DGM4, and 74.9 F1 on MMFakeBench.
Authors: Hari Prasanth S. M., Pejman Habibiroudkenar, Eerik Alamikkotervo, Dimitrios Bouzoulas, Risto Ojala
Abstract: Event cameras provide high-temporal-resolution visual sensing that is well suited for observing fast-moving aerial objects; however, their use for drone trajectory prediction remains limited. This work introduces an event-only drone forecasting method that exploits propeller-induced motion cues. Propeller rotational speed are extracted directly from raw event data and fused within an RPM-aware Kalman filtering framework. Evaluations on the FRED dataset show that the proposed method outperforms learning-based approaches and vanilla kalman filter in terms of average distance error and final distance error at 0.4s and 0.8s forecasting horizons. The results demonstrate robust and accurate short- and medium-horizon trajectory forecasting without reliance on RGB imagery or training data.
Authors: Peiyuan Jing, Chun-Wun Cheng, Liutao Yang, Zhenxuan Zhang, Thiago V. Lima, Klaus Strobel, Antoine Leimgruber, Angelica Aviles-Rivero, Guang Yang, Javier A. Montoya-Zegarra
Abstract: Low-dose Positron Emission Tomography (PET) reduces radiation exposure but suffers from severe noise and quantitative degradation. Diffusion-based denoising models achieve strong final reconstructions, yet their reverse trajectories are typically unconstrained and not aligned with the progressive nature of PET dose formation. We propose MAP-Diff, a multi-anchor guided diffusion framework for progressive 3D whole-body PET denoising. MAP-Diff introduces clinically observed intermediate-dose scans as trajectory anchors and enforces timestep-dependent supervision to regularize the reverse process toward dose-aligned intermediate states. Anchor timesteps are calibrated via degradation matching between simulated diffusion corruption and real multi-dose PET pairs, and a timestep-weighted anchor loss stabilizes stage-wise learning. At inference, the model requires only ultra-low-dose input while enabling progressive, dose-consistent intermediate restoration. Experiments on internal (Siemens Biograph Vision Quadra) and cross-scanner (United Imaging uEXPLORER) datasets show consistent improvements over strong CNN-, Transformer-, GAN-, and diffusion-based baselines. On the internal dataset, MAP-Diff improves PSNR from 42.48 dB to 43.71 dB (+1.23 dB), increases SSIM to 0.986, and reduces NMAE from 0.115 to 0.103 (-0.012) compared to 3D DDPM. Performance gains generalize across scanners, achieving 34.42 dB PSNR and 0.141 NMAE on the external cohort, outperforming all competing methods.
Authors: Simon Ging (Computer Vision Group, University of Freiburg, Germany, Adaptive & Agentic AI), Philipp Arnold (Department of Radiology, Medical Center -- University of Freiburg, Germany), Sebastian Walter (Chair of Algorithms and Data Structures, University of Freiburg, Germany), Hani Alnahas (Computer Vision Group, University of Freiburg, Germany), Hannah Bast (Chair of Algorithms and Data Structures, University of Freiburg, Germany), Elmar Kotter (Department of Radiology, Medical Center -- University of Freiburg, Germany), Jiancheng Yang (ELLIS Institute Finland, School of Electrical Engineering, Aalto University, Finland), Behzad Bozorgtabar (Adaptive & Agentic AI), Thomas Brox (Computer Vision Group, University of Freiburg, Germany)
Abstract: Recent 3D CT vision-language models align volumes with reports via contrastive pretraining, but typically rely on limited public data and provide only coarse global supervision. We train a 3D CT vision-language model on 98k report-volume pairs (50k patients) collected at a single hospital, combined with public datasets, using SigLIP-style contrastive pretraining together with prompt-based disease supervision in the shared vision-text embedding space. On CT-RATE, our model achieves state-of-the-art text-to-image retrieval (R@10 31.5 vs. 22.2) and competitive disease classification (AUC 83.8 vs. 83.8), with consistent results on Rad-ChestCT (AUC 77.0 vs. 77.3). We further observe that radiologists routinely reference specific images within their reports (e.g., ``series X, image Y''), linking textual descriptions to precise axial locations. We automatically mine 262k such snippet-slice pairs and introduce the task of intra-scan snippet localization -- predicting the axial depth referred to by a text snippet -- reducing mean absolute error to 36.3 mm at 12 mm feature resolution, compared with 67.0 mm for the best baseline. Adding this localization objective leaves retrieval and classification broadly unchanged within confidence bounds, yielding a single unified model for retrieval, classification, and intra-scan grounding.
Authors: Manuel Serna-Aguilera, Raegan Anderes, Page Dobbs, Khoa Luu
Abstract: The nicotine addiction public health crisis continues to be pervasive. In this century alone, the tobacco industry has released and marketed new products in an aggressive effort to lure new and young customers for life. Such innovations and product development, namely flavored nicotine or tobacco such as nicotine pouches, have undone years of anti-tobacco campaign work. Past work is limited both in scope and in its ability to connect large-scale data points. Thus, we introduce the Nicotine Innovation Counter-Offensive (NICO) Dataset to provide public health researchers with over 200,000 multimodal samples, including images and text descriptions, on 55 tobacco and nicotine product brands. In addition, to provide public health researchers with factual connections across a large-scale dataset, we propose NICO-RAG, a retrieval-augmented generation (RAG) framework that can retrieve image features without incurring the high-cost of language models, as well as the added cost of processing image tokens with large-scale datasets such as NICO. At construction time, NICO-RAG organizes image- and text-extracted entities and relations into hypergraphs to produce as factual responses as possible. This joint multimodal knowledge representation enables NICO-RAG to retrieve images for query answering not only by visual similarity but also by the semantic similarity of image descriptions. Experimentals show that without needing to process additional tokens from images for over 100 questions, NICO-RAG performs comparably to the state-of-the-art RAG method adapted for images.
Authors: Yisu Zhang, Chenjie Cao, Tengfei Wang, Xuhui Zuo, Junta Wu, Jianke Zhu, Chunchao Guo
Abstract: Recent advances in foundational Video Diffusion Models (VDMs) have yielded significant progress. Yet, despite the remarkable visual quality of generated videos, reconstructing consistent 3D scenes from these outputs remains challenging, due to limited camera controllability and inconsistent generated content when viewed from distinct camera trajectories. In this paper, we propose WorldStereo, a novel framework that bridges camera-guided video generation and 3D reconstruction via two dedicated geometric memory modules. Formally, the global-geometric memory enables precise camera control while injecting coarse structural priors through incrementally updated point clouds. Moreover, the spatial-stereo memory constrains the model's attention receptive fields with 3D correspondence to focus on fine-grained details from the memory bank. These components enable WorldStereo to generate multi-view-consistent videos under precise camera control, facilitating high-quality 3D reconstruction. Furthermore, the flexible control branch-based WorldStereo shows impressive efficiency, benefiting from the distribution matching distilled VDM backbone without joint training. Extensive experiments across both camera-guided video generation and 3D reconstruction benchmarks demonstrate the effectiveness of our approach. Notably, we show that WorldStereo acts as a powerful world model, tackling diverse scene generation tasks (whether starting from perspective or panoramic images) with high-fidelity 3D results. Models will be released.
Authors: Jo\"el K\"uchler, Ellen van Maren, Vaiva Vasiliauskait\.e, Katarina Vuli\'c, Reza Abbasi-Asl, Stephan J. Ihle
Abstract: Although data generation is often straightforward, extracting information from data is more difficult. Object-centric representation learning can extract information from images in an unsupervised manner. It does so by segmenting an image into its subcomponents: the objects. Each object is then represented in a low-dimensional latent space that can be used for downstream processing. Object-centric representation learning is dominated by autoencoder architectures (AEs). Here, we present ORGAN, a novel approach for object-centric representation learning, which is based on cycle-consistent Generative Adversarial Networks instead. We show that it performs similarly to other state-of-the-art approaches on synthetic datasets, while at the same time being the only approach tested here capable of handling more challenging real-world datasets with many objects and low visual contrast. Complementing these results, ORGAN creates expressive latent space representations that allow for object manipulation. Finally, we show that ORGAN scales well both with respect to the number of objects and the size of the images, giving it a unique edge over current state-of-the-art approaches.
Authors: Zhengyang Xu, Han Li, Jingsong Liu, Linrui Xie, Xun Ma, Xin You, Shihui Zu, Ayako Ito, Xinyu Hao, Hongming Xu, Shaohua Kevin Zhou, Nassir Navab, Peter J. Sch\"uffler
Abstract: Recent AI navigation approaches aim to improve Whole-Slide Image (WSI) diagnosis by modeling spatial exploration and selecting diagnostically relevant regions, yet most operate at a single fixed magnification or rely on predefined magnification traversal. In clinical practice, pathologists examine slides across multiple magnifications and selectively inspect only necessary scales, dynamically integrating global and cellular evidence in a sequential manner. This mismatch prevents existing methods from modeling cross-magnification interactions and adaptive magnification selection inherent to real diagnostic workflows. To these, we propose a clinically consistent Multi-Magnification WSI Navigation Agent (MMNavAgent) that explicitly models multi magnification interaction and adaptive magnification selection. Specifically, we introduce a Cross-Magnification navigation Tool (CMT) that aggregates contextual information from adjacent magnifications to enhance discriminative representations along the navigation path. We further introduce a Magnification Selection Tool (MST) that leverages memory-driven reasoning within the agent framework to enable interactive and adaptive magnification selection, mimicking the sequential decision process of pathologists. Extensive experiments on a public dataset demonstrate improved diagnostic performance, with 1.45% gain of AUC and 2.93% gain of BACC over a non-agent baseline. Code will be public upon acceptance.
Authors: Isaac Corley, Caleb Robinson, Inbal Becker-Reshef, Juan M. Lavista Ferres
Abstract: As geospatial foundation models shift from patch-level to pixel-level embeddings, practitioners must aggregate thousands of pixel vectors into patch representations that preserve class-discriminative signal while matching downstream label resolution. The default choice, mean pooling, discards within-patch variability and can drop accuracy by more than 10% under spatial shift. To evaluate this effect, we introduce EuroSAT-Embed: 81,000 embedding GeoTIFFs derived from three foundation models: AlphaEarth, OlmoEarth, and Tessera. We benchmark 11 training-free and 2 parametric pooling methods under both random and geographically disjoint test splits. Our results show that richer pooling schemes reduce the geographic generalization gap by up to 40% relative to mean pooling and increases accuracy by up to 5% on spatial splits. We recommend Generalized Mean Pooling (GeM) as a drop-in replacement for mean pooling: it improves accuracy without increasing embedding dimensionality. For maximum accuracy, Stats pooling (concatenation of min/max/mean/std pooling) performs best at 4x the embedding size. We further find that pooling effectiveness varies across embedding sources and that higher-dimensional embeddings benefit most from distributional statistics.
Authors: Harikrishnan Unnikrishnan
Abstract: Background: Accurate glottal segmentation in high-speed videoendoscopy (HSV) is essential for extracting kinematic biomarkers of laryngeal function. However, existing deep learning models often produce spurious artifacts in non-glottal frames and fail to generalize across different clinical settings. Methods: We propose a detection-gated pipeline that integrates a YOLOv8-based detector with a U-Net segmenter. A temporal consistency wrapper ensures robustness by suppressing false positives during glottal closure and instrument occlusion. The model was trained on a limited subset of the GIRAFE dataset (600 frames) and evaluated via zero-shot transfer on the large-scale BAGLS dataset. Results: The pipeline achieved state-of-the-art performance on the GIRAFE benchmark (DSC 0.81) and demonstrated superior generalizability on BAGLS (DSC 0.85, in-distribution) without institutional fine-tuning. Downstream validation on a 65-subject clinical cohort confirmed that automated kinematic features (Open Quotient, coefficient of variation) remained consistent with established clinical benchmarks. The coefficient of variation (CV) of the glottal area was found to be a significant marker for distinguishing healthy from pathological vocal function (p=0.006). Conclusions: The detection-gated architecture provides a lightweight, computationally efficient solution (~35 frames/s) for real-time clinical use. By enabling robust zero-shot transfer, this framework facilitates the standardized, large-scale extraction of clinical biomarkers across diverse endoscopy platforms. Code, trained weights, and evaluation scripts are released at https://github.com/hari-krishnan/openglottal.
Authors: Yiweng Xie, Bo He, Junke Wang, Xiangyu Zheng, Ziyi Ye, Zuxuan Wu
Abstract: This paper presents FluxMem, a training-free framework for efficient streaming video understanding. FluxMem adaptively compresses redundant visual memory through a hierarchical, two-stage design: (1) a Temporal Adjacency Selection (TAS) module removes redundant visual tokens across adjacent frames, and (2) a Spatial Domain Consolidation (SDC) module further merges spatially repetitive regions within each frame into compact representations. To adapt effectively to dynamic scenes, we introduce a self-adaptive token compression mechanism in both TAS and SDC, which automatically determines the compression rate based on intrinsic scene statistics rather than manual tuning. Extensive experiments demonstrate that FluxMem achieves new state-of-the-art results on existing online video benchmarks, reaching 76.4 on StreamingBench and 67.2 on OVO-Bench under real-time settings, while reducing latency by 69.9% and peak GPU memory by 34.5% on OVO-Bench. Furthermore, it maintains strong offline performance, achieving 73.1 on MLVU while using 65% fewer visual tokens.
Authors: Germain Bregeon, Marius Preda, Radu Ispas, Titus Zaharia
Abstract: In this paper, we introduce a novel 3D mesh convolution-based autoencoder for geometry compression, able to deal with irregular mesh data without requiring neither preprocessing nor manifold/watertightness conditions. The proposed approach extracts meaningful latent representations by learning features directly from the mesh faces, while preserving connectivity through dedicated pooling and unpooling operations. The encoder compresses the input mesh into a compact base mesh space, which ensures that the latent space remains comparable. The decoder reconstructs the original connectivity and restores the compressed geometry to its full resolution. Extensive experiments on multi-class datasets demonstrate that our method outperforms state-of-the-art approaches in both 3D mesh geometry reconstruction and latent space classification tasks. Code available at: github.com/germainGB/MeshConv3D
Authors: Hualiang Wei, Shunran Jia, Jialun Liu, Wenhui Li
Abstract: We present LiftAvatar, a new paradigm that completes sparse monocular observations in kinematic space (e.g., facial expressions and head pose) and uses the completed signals to drive high-fidelity avatar animation. LiftAvatar is a fine-grained, expression-controllable large-scale video diffusion Transformer that synthesizes high-quality, temporally coherent expression sequences conditioned on single or multiple reference images. The key idea is to lift incomplete input data into a richer kinematic representation, thereby strengthening both reconstruction and animation in downstream 3D avatar pipelines. To this end, we introduce (i) a multi-granularity expression control scheme that combines shading maps with expression coefficients for precise and stable driving, and (ii) a multi-reference conditioning mechanism that aggregates complementary cues from multiple frames, enabling strong 3D consistency and controllability. As a plug-and-play enhancer, LiftAvatar directly addresses the limited expressiveness and reconstruction artifacts of 3D Gaussian Splatting-based avatars caused by sparse kinematic cues in everyday monocular videos. By expanding incomplete observations into diverse pose-expression variations, LiftAvatar also enables effective prior distillation from large-scale video generative models into 3D pipelines, leading to substantial gains. Extensive experiments show that LiftAvatar consistently boosts animation quality and quantitative metrics of state-of-the-art 3D avatar methods, especially under extreme, unseen expressions.
Authors: Tutian Tang, Xingyu Ji, Yutong Li, MingHao Liu, Wenqiang Xu, Cewu Lu
Abstract: Recent advancements in visual-inertial motion capture systems have demonstrated the potential of combining monocular cameras with sparse inertial measurement units (IMUs) as cost-effective solutions, which effectively mitigate occlusion and drift issues inherent in single-modality systems. However, they are still limited by metric inaccuracies in global translations stemming from monocular depth ambiguity, and shape-agnostic local motion estimations that ignore anthropometric variations. We present Stereo-Inertial Poser, a real-time motion capture system that leverages a single stereo camera and six IMUs to estimate metric-accurate and shape-aware 3D human motion. By replacing the monocular RGB with stereo vision, our system resolves depth ambiguity through calibrated baseline geometry, enabling direct 3D keypoint extraction and body shape parameter estimation. IMU data and visual cues are fused for predicting drift-compensated joint positions and root movements, while a novel shape-aware fusion module dynamically harmonizes anthropometry variations with global translations. Our end-to-end pipeline achieves over 200 FPS without optimization-based post-processing, enabling real-time deployment. Quantitative evaluations across various datasets demonstrate state-of-the-art performance. Qualitative results show our method produces drift-free global translation under a long recording time and reduces foot-skating effects.
Authors: Chong Xia, Kai Zhu, Zizhuo Wang, Fangfu Liu, Zhizheng Zhang, Yueqi Duan
Abstract: Compositional scene reconstruction seeks to create object-centric representations rather than holistic scenes from real-world videos, which is natively applicable for simulation and interaction. Conventional compositional reconstruction approaches primarily emphasize on visual appearance and show limited generalization ability to real-world scenarios. In this paper, we propose SimRecon, a framework that realizes a "Perception-Generation-Simulation" pipeline towards cluttered scene reconstruction, which first conducts scene-level semantic reconstruction from video input, then performs single-object generation, and finally assembles these assets in the simulator. However, naively combining these three stages leads to visual infidelity of generated assets and physical implausibility of the final scene, a problem particularly severe for complex scenes. Thus, we further propose two bridging modules between the three stages to address this problem. To be specific, for the transition from Perception to Generation, critical for visual fidelity, we introduce Active Viewpoint Optimization, which actively searches in 3D space to acquire optimal projected images as conditions for single-object completion. Moreover, for the transition from Generation to Simulation, essential for physical plausibility, we propose a Scene Graph Synthesizer, which guides the construction from scratch in 3D simulators, mirroring the native, constructive principle of the real world. Extensive experiments on the ScanNet dataset validate our method's superior performance over previous state-of-the-art approaches.
Authors: Chong Xia, Fangfu Liu, Yule Wang, Yize Pang, Yueqi Duan
Abstract: Recent advances in generalizable 3D Gaussian Splatting (3DGS) have enabled rapid 3D scene reconstruction within seconds, eliminating the need for per-scene optimization. However, existing methods primarily follow an offline reconstruction paradigm, lacking the capacity for continuous reconstruction, which limits their applicability to online scenarios such as robotics and VR/AR. In this paper, we introduce OnlineX, a feed-forward framework that reconstructs both 3D visual appearance and language fields in an online manner using only streaming images. A key challenge in online formulation is the cumulative drift issue, which is rooted in the fundamental conflict between two opposing roles of the memory state: an active role that constantly refreshes to capture high-frequency local geometry, and a stable role that conservatively accumulates and preserves the long-term global structure. To address this, we introduce a decoupled active-to-stable state evolution paradigm. Our framework decouples the memory state into a dedicated active state and a persistent stable state, and then cohesively fuses the information from the former into the latter to achieve both fidelity and stability. Moreover, we jointly model visual appearance and language fields and incorporate an implicit Gaussian fusion module to enhance reconstruction quality. Experiments on mainstream datasets demonstrate that our method consistently outperforms prior work in novel view synthesis and semantic understanding, showcasing robust performance across input sequences of varying lengths with real-time inference speed.
Authors: Yiying Yang, Wei Cheng, Sijin Chen, Honghao Fu, Xianfang Zeng, Yujun Cai, Gang Yu, Xingjun Ma
Abstract: OmniLottie is a versatile framework that generates high quality vector animations from multi-modal instructions. For flexible motion and visual content control, we focus on Lottie, a light weight JSON formatting for both shapes and animation behaviors representation. However, the raw Lottie JSON files contain extensive invariant structural metadata and formatting tokens, posing significant challenges for learning vector animation generation. Therefore, we introduce a well designed Lottie tokenizer that transforms JSON files into structured sequences of commands and parameters representing shapes, animation functions and control parameters. Such tokenizer enables us to build OmniLottie upon pretrained vision language models to follow multi-modal interleaved instructions and generate high quality vector animations. To further advance research in vector animation generation, we curate MMLottie-2M, a large scale dataset of professionally designed vector animations paired with textual and visual annotations. With extensive experiments, we validate that OmniLottie can produce vivid and semantically aligned vector animations that adhere closely to multi modal human instructions.
Authors: Kwame Mbobda-Kuate, Gabriel Kasmi
Abstract: Scaling laws assume larger models trained on more data consistently outperform smaller ones -- an assumption that drives model selection in computer vision but remains untested in resource-constrained Earth observation (EO). We conduct a systematic efficiency analysis across three scaling dimensions: model size, dataset size, and input resolution, on rooftop PV detection in Madagascar. Optimizing for model efficiency (mAP$_{50}$ per unit of model size), we find a consistent efficiency inversion: YOLO11N achieves both the highest efficiency ($24\times$ higher than YOLO11X) and the highest absolute mAP$_{50}$ (0.617). Resolution is the dominant resource allocation lever ($+$120% efficiency gain), while additional data yields negligible returns at low resolution. These findings are robust to the deployment objective: small high-resolution configurations are Pareto-dominant across all 44 setups in the joint accuracy-throughput space, leaving no tradeoff to resolve. In data-scarce EO, bigger is not just unnecessary: it can be worse.
Authors: Namhoon Kim, Narges Moeini, Justin Romberg, Sara Fridovich-Keil
Abstract: Volume denoising is a foundational problem in computational imaging, as many 3D imaging inverse problems face high levels of measurement noise. Inspired by the strong 2D image denoising properties of Field of Junctions (ICCV 2021), we propose a novel, fully volumetric 3D Field of Junctions (3D FoJ) representation that optimizes a junction of 3D wedges that best explain each 3D patch of a full volume, while encouraging consistency between overlapping patches. In addition to direct volume denoising, we leverage our 3D FoJ representation as a structural prior that: (i) requires no training data, and thus precludes the risk of hallucination, (ii) preserves and enhances sharp edge and corner structures in 3D, even under low signal to noise ratio (SNR), and (iii) can be used as a drop-in denoising representation via projected or proximal gradient descent for any volumetric inverse problem with low SNR. We demonstrate successful volume reconstruction and denoising with 3D FoJ across three diverse 3D imaging tasks with low-SNR measurements: low-dose X-ray computed tomography (CT), cryogenic electron tomography (cryo-ET), and denoising point clouds such as those from lidar in adverse weather. Across these challenging low-SNR volumetric imaging problems, 3D FoJ outperforms a mixture of classical and neural methods.
Authors: Aniek Eijpe, Soufyan Lakbir, Melis Erdal Cesur, Sara P. Oliveira, Angelos Chatzimparmpas, Sanne Abeln, Wilson Silva
Abstract: While multimodal survival prediction models are increasingly more accurate, their complexity often reduces interpretability, limiting insight into how different data sources influence predictions. To address this, we introduce DIMAFx, an explainable multimodal framework for cancer survival prediction that produces disentangled, interpretable modality-specific and modality-shared representations from histopathology whole-slide images and transcriptomics data. Across multiple cancer cohorts, DIMAFx achieves state-of-the-art performance and improved representation disentanglement. Leveraging its interpretable design and SHapley Additive exPlanations, DIMAFx systematically reveals key multimodal interactions and the biological information encoded in the disentangled representations. In breast cancer survival prediction, the most predictive features contain modality-shared information, including one capturing solid tumor morphology contextualized primarily by late estrogen response, where higher-grade morphology aligned with pathway upregulation and increased risk, consistent with known breast cancer biology. Key modality-specific features capture microenvironmental signals from interacting adipose and stromal morphologies. These results show that multimodal models can overcome the traditional trade-off between performance and explainability, supporting their application in precision medicine.
Authors: Srikumar Sastry, Dan Cher, Brian Wei, Aayush Dhakal, Subash Khanal, Dev Gupta, Nathan Jacobs
Abstract: We introduce GeoDiT, a diffusion transformer designed for text-to-satellite image generation with point-based control. Existing controlled satellite image generative models often require pixel-level maps that are time-consuming to acquire, yet semantically limited. To address this limitation, we introduce a novel point-based conditioning framework that controls the generation process through the spatial location of the points and the textual description associated with each point, providing semantically rich control signals. This approach enables flexible, annotation-friendly, and computationally simple inference for satellite image generation. To this end, we introduce an adaptive local attention mechanism that effectively regularizes the attention scores based on the input point queries. We systematically evaluate various domain-specific design choices for training GeoDiT, including the selection of satellite image representation for alignment and geolocation representation for conditioning. Our experiments demonstrate that GeoDiT achieves impressive generation performance, surpassing the state-of-the-art remote sensing generative models.
Authors: Yiqi Lin, Guoqiang Liang, Ziyun Zeng, Zechen Bai, Yanzhe Chen, Mike Zheng Shou
Abstract: Instruction-based video editing has witnessed rapid progress, yet current methods often struggle with precise visual control, as natural language is inherently limited in describing complex visual nuances. Although reference-guided editing offers a robust solution, its potential is currently bottlenecked by the scarcity of high-quality paired training data. To bridge this gap, we introduce a scalable data generation pipeline that transforms existing video editing pairs into high-fidelity training quadruplets, leveraging image generative models to create synthesized reference scaffolds. Using this pipeline, we construct RefVIE, a large-scale dataset tailored for instruction-reference-following tasks, and establish RefVIE-Bench for comprehensive evaluation. Furthermore, we propose a unified editing architecture, Kiwi-Edit, that synergizes learnable queries and latent visual features for reference semantic guidance. Our model achieves significant gains in instruction following and reference fidelity via a progressive multi-stage training curriculum. Extensive experiments demonstrate that our data and architecture establish a new state-of-the-art in controllable video editing. All datasets, models, and code is released at https://github.com/showlab/Kiwi-Edit.
Authors: Quoc-Khang Tran, Minh-Thien Nguyen, Nguyen-Khang Pham
Abstract: The classification of Intangible Cultural Heritage (ICH) images in the Mekong Delta poses unique challenges due to limited annotated data, high visual similarity among classes, and domain heterogeneity. In such low-resource settings, conventional deep learning models often suffer from high variance or overfit to spurious correlations, leading to poor generalization. To address these limitations, we propose a robust framework that integrates the hybrid CoAtNet architecture with model soups, a lightweight weight-space ensembling technique that averages checkpoints from a single training trajectory without increasing inference cost. CoAtNet captures both local and global patterns through stage-wise fusion of convolution and self-attention. We apply two ensembling strategies - greedy and uniform soup - to selectively combine diverse checkpoints into a final model. Beyond performance improvements, we analyze the ensembling effect through the lens of bias-variance decomposition. Our findings show that model soups reduces variance by stabilizing predictions across diverse model snapshots, while introducing minimal additional bias. Furthermore, using cross-entropy-based distance metrics and Multidimensional Scaling (MDS), we show that model soups selects geometrically diverse checkpoints, unlike Soft Voting, which blends redundant models centered in output space. Evaluated on the ICH-17 dataset (7,406 images across 17 classes), our approach achieves state-of-the-art results with 72.36% top-1 accuracy and 69.28% macro F1-score, outperforming strong baselines including ResNet-50, DenseNet-121, and ViT. These results underscore that diversity-aware checkpoint averaging provides a principled and efficient way to reduce variance and enhance generalization in culturally rich, data-scarce classification tasks.
Authors: Divyanshu Daiya, Aniket Bera
Abstract: We present Sketch2Colab, which turns storyboard-style 2D sketches into coherent, object-aware 3D multi-human motion with fine-grained control over agents, joints, timing, and contacts. Conventional diffusion-based motion generators have advanced realism; however, achieving precise adherence to rich interaction constraints typically demands extensive training and/or costly posterior guidance, and performance can degrade under strong multi-entity conditioning. Sketch2Colab instead first learns a sketch-driven diffusion prior and then distills it into an efficient rectified-flow student operating in latent space for fast, stable sampling. Differentiable energies over keyframes, trajectories, and physics-based constraints directly shape the student's transport field, steering samples toward motions that faithfully satisfy the storyboard while remaining physically plausible. To capture coordinated interaction, we augment the continuous flow with a continuous-time Markov chain (CTMC) planner that schedules discrete events such as touches, grasps, and handoffs, modulating the dynamics to produce crisp, well-phased human-object-human collaborations. Experiments on CORE4D and InterHuman show that Sketch2Colab achieves state-of-the-art constraint adherence and perceptual quality while offering significantly faster inference than diffusion-only baselines.
Authors: Mateus Karvat, Bram Adams, Sidney Givigi
Abstract: Autonomous vehicle (AV) perception models are typically evaluated solely on benchmark performance metrics, with limited attention to code quality, production readiness and long-term maintainability. This creates a significant gap between research excellence and real-world deployment in safety-critical systems subject to international safety standards. To address this gap, we present the first large-scale empirical study of software quality in AV perception repositories, systematically analyzing 178 unique models from the KITTI and NuScenes 3D Object Detection leaderboards. Using static analysis tools (Pylint, Bandit, and Radon), we evaluated code errors, security vulnerabilities, maintainability, and development practices. Our findings revealed that only 7.3% of the studied repositories meet basic production-readiness criteria, defined as having zero critical errors and no high-severity security vulnerabilities. Security issues are highly concentrated, with the top five issues responsible for almost 80% of occurrences, which prompted us to develop a set of actionable guidelines to prevent them. Additionally, the adoption of Continuous Integration/Continuous Deployment pipelines was correlated with better code maintainability. Our findings highlight that leaderboard performance does not reflect production readiness and that targeted interventions could substantially improve the quality and safety of AV perception code.
Authors: Moru Liu, Hao Dong, Olga Fink, Mario Trapp
Abstract: The deployment of multimodal models in high-stakes domains, such as self-driving vehicles and medical diagnostics, demands not only strong predictive performance but also reliable mechanisms for detecting failures. In this work, we address the largely unexplored problem of failure detection in multimodal contexts. We propose Adaptive Confidence Regularization (ACR), a novel framework specifically designed to detect multimodal failures. Our approach is driven by a key observation: in most failure cases, the confidence of the multimodal prediction is significantly lower than that of at least one unimodal branch, a phenomenon we term confidence degradation. To mitigate this, we introduce an Adaptive Confidence Loss that penalizes such degradations during training. In addition, we propose Multimodal Feature Swapping, a novel outlier synthesis technique that generates challenging, failure-aware training examples. By training with these synthetic failures, ACR learns to more effectively recognize and reject uncertain predictions, thereby improving overall reliability. Extensive experiments across four datasets, three modalities, and multiple evaluation settings demonstrate that ACR achieves consistent and robust gains. The source code will be available at https://github.com/mona4399/ACR.
Authors: Yichen Liu, Donghao Zhou, Jie Wang, Xin Gao, Guisheng Liu, Jiatong Li, Quanwei Zhang, Qiang Lyu, Lanqing Guo, Shilei Wen, Weiqiang Wang, Pheng-Ann Heng
Abstract: Human-product images, which showcase the integration of humans and products, play a vital role in advertising, e-commerce, and digital marketing. The essential challenge of generating such images lies in ensuring the high-fidelity preservation of product details. Among existing paradigms, reference-based inpainting offers a targeted solution by leveraging product reference images to guide the inpainting process. However, limitations remain in three key aspects: the lack of diverse large-scale training data, the struggle of current models to focus on product detail preservation, and the inability of coarse supervision for achieving precise guidance. To address these issues, we propose HiFi-Inpaint, a novel high-fidelity reference-based inpainting framework tailored for generating human-product images. HiFi-Inpaint introduces Shared Enhancement Attention (SEA) to refine fine-grained product features and Detail-Aware Loss (DAL) to enforce precise pixel-level supervision using high-frequency maps. Additionally, we construct a new dataset, HP-Image-40K, with samples curated from self-synthesis data and processed with automatic filtering. Experimental results show that HiFi-Inpaint achieves state-of-the-art performance, delivering detail-preserving human-product images.
Authors: Mohamed Bayan Kmainasi, Abul Hasnat, Md Arid Hasan, Ali Ezzat Shahroor, Firoj Alam
Abstract: The proliferation of multimodal content on social media presents significant challenges in understanding and moderating complex, context-dependent issues such as misinformation, hate speech, and propaganda. While efforts have been made to develop resources and propose new methods for automatic detection, limited attention has been given to jointly modeling label detection and the generation of explanation-based rationales, which often leads to degraded classification performance when trained simultaneously. To address this challenge, we introduce MemeXplain, an explanation-enhanced dataset for propagandistic memes in Arabic and hateful memes in English, making it the first large-scale resource for these tasks. To solve these tasks, we propose a multi-stage optimization approach and train Vision-Language Models (VLMs). Our results show that this strategy significantly improves both label detection and explanation generation quality over the base model, outperforming the current state-of-the-art with an absolute improvement of ~1.4% (Acc) on ArMeme and ~2.2% (Acc) on Hateful Memes. For reproducibility and future research, we aim to make the MemeXplain dataset and scripts publicly available (https://github.com/MohamedBayan/MemeIntel).
Authors: Datorien L. Anderson
Abstract: Standard evaluation metrics for machine learning -- accuracy, precision, recall, and AUROC -- assume that all errors are equivalent: a confident incorrect prediction is penalized identically to an uncertain one. For discrete commitment systems (architectures that select committed states {-W, 0, +W}), this assumption is epistemologically flawed. We introduce the Certainty-Validity (CVS) Framework, a diagnostic method that decomposes model performance into a 2x2 matrix distinguishing high/low certainty from valid/invalid predictions. This framework reveals a critical failure mode hidden by standard accuracy: Confident-Incorrect (CI) behavior, where models hallucinate structure in ambiguous data. Through ablation experiments on Fashion-MNIST, EMNIST, and IMDB, we analyze the "83% Ambiguity Ceiling" -- a stopping point where this specific discrete architecture consistently plateaus on noisy benchmarks. Unlike continuous models that can surpass this ceiling by memorizing texture or statistical noise, the discrete model refuses to commit to ambiguous samples. We show that this refusal is not a failure but a feature: the model stops where structural evidence ends. However, standard training on ambiguous data eventually forces Benign Overfitting, causing a pathological migration from Uncertain-Incorrect (appropriate doubt) to Confident-Incorrect (hallucination). We propose that "good training" for reasoning systems must be defined not by accuracy, but by maximizing the Certainty-Validity Score (CVS) -- ensuring the model knows where to stop.
Authors: Zhen Peng, Peter J. Bentley
Abstract: Accurate evaluation of building energy performance remains challenging in regions where scalable Energy Performance Certificate (EPC) assessments are unavailable. This paper presents a cost-efficient framework that leverages Vision-Language models for automated EPC pre-assessment from limited visual information. The proposed Multimodal Modular Chain of Thoughts (MMCoT) architecture decomposes EPC estimation into intermediate reasoning stages and explicitly propagates inferred attributes across tasks using structured prompting. Experiments on a multimodal dataset of 81 residential properties in the United Kingdom show that MMCoT achieves statistically significant improvements over instruction-only prompting for EPC estimation. Analysis based on accuracy, recall, mean absolute error, and confusion matrices indicate that the proposed approach captures the ordinal structure of EPC ratings, with most errors occurring between adjacent classes. These results suggest that modular prompt-based reasoning offers a promising direction for low-cost EPC pre-assessment in data-scarce settings.
Authors: Elena Zoppellari, Federico Becattini, Marco Fiorucci, Lamberto Ballan
Abstract: For robots to operate effectively and safely alongside humans, they must be able to understand the progress of ongoing actions. This ability, known as action progress prediction, is critical for tasks ranging from timely assistance to autonomous decision-making. However, modeling action progression in robotics has often been overlooked. Moreover, a single camera may be insufficient for understanding robot's ego-actions, as self-occlusion can significantly hinder perception and model performance. In this paper, we propose a multi-view architecture for action progress prediction in robot manipulation tasks. Experiments on Mobile ALOHA demonstrate the effectiveness of the proposed approach.
Authors: Joy T Wu, Daniel Beckmann, Sarah Miller, Alexander Lee, Elizabeth Theng, Stephan Altmayer, Ken Chang, David Kersting, Tomoaki Otani, Brittany Z Dashevsky, Hye Lim Park, Matteo Novello, Kip Guja, Curtis Langlotz, Ismini Lourentzou, Daniel Gruhl, Benjamin Risse, Guido A Davidzon
Abstract: [18F]FDG-PET/CT is a cornerstone imaging modality for tumor staging and treatment response assessment across many cancer types, yet expert reader shortages necessitate more efficient diagnostic aids. While standalone AI models for automatic lesion segmentation exist, clinical translation remains hindered by concerns about interpretability, explainability, reliability, and workflow integration. We present GazeXPErT, a 4D eye-tracking dataset capturing expert search patterns during tumor detection and measurement on 346 FDG-PET/CT scans. Each study was read by a trainee and a board-certified nuclear medicine or radiology specialist using an eye-tracking-enabled annotation platform that simulates routine clinical reads. From 3,948 minutes of raw 60Hz eye-tracking data, 9,030 unique gaze-to-lesion trajectories were extracted, synchronized with PET/CT image slices, and rendered in COCO-style format for multiple machine learning applications. Baseline validation experiments demonstrate that a 3D nnUNet tumor segmentation model achieved superior performance when incorporating expert gaze patterns versus without (DICE score 0.6819 versus 0.6008), and that vision transformers trained on sequential gaze and PET/CT images can improve dynamic lesion localization (74.95% predicted gaze point closer to tumor) and expert intention prediction (Accuracy 67.53% and AUROC 0.747). GazeXPErT is a valuable resource designed to explore multiple machine learning problems beyond these baseline experiments, which include and are not limited to, visual grounding or causal reasoning, clinically explainable feature augmentation, human-computer interaction, human intention prediction or understanding, and expert gaze-rewarded modeling approaches to AI in oncologic FDG-PET/CT imaging.
Authors: Lingfeng He, De Cheng, Huaijie Wang, Xi Yang, Nannan Wang, Xinbo Gao
Abstract: Continual Learning (CL) requires models to sequentially adapt to new tasks without forgetting old knowledge. Recently, Low-Rank Adaptation (LoRA), a representative Parameter-Efficient Fine-Tuning (PEFT) method, has gained increasing attention in CL. Several LoRA-based CL methods reduce interference across tasks by separating their update spaces, typically building the new space from the estimated null space of past tasks. However, they (i) overlook task-shared directions, which suppresses knowledge transfer, and (ii) fail to capture truly effective task-specific directions since these ``null bases" of old tasks can remain nearly inactive for new task under correlated tasks. To address this, we study LoRA learning capability from a projection energy perspective, and propose Low-rank Decomposition and Adaptation (LoDA). It performs a task-driven decomposition to build general and truly task-specific LoRA subspaces by solving two energy-based objectives, decoupling directions for knowledge sharing and isolation. LoDA fixes LoRA down-projections on two subspaces and learns robust up-projections via a Gradient-Aligned Optimization (GAO) approach. After each task, before integrating the LoRA updates into the backbone, LoDA derives a closed-form recalibration for the general update, approximating a feature-level joint optimum along this task-shared direction. Experiments indicate that LoDA outperforms existing CL methods.
Authors: Muneeba Rashid, Hina Shakir, Humaira Mehwish, Asarim Amir, Reema Qaiser Khan
Abstract: Magnetic Resonance (MR) imaging is a diagnostic tool used in modern medicine; however, its output can be affected by motion artefacts and may be limited by equipment. This research focuses on MRI image quality enhancement using two efficient Generative Adversarial Networks (GANs) models: SOUP-GAN and CSR-GAN. In both models, meaningful architectural modifications were introduced. The generator and discriminator of each were further deepened by adding convolutional layers and were enhanced in filter sizes as well. The LeakyReLU activation function was used to improve gradient flow, and hyperparameter tuning strategies were applied, including a reduced learning rate and an optimal batch size. Moreover, spectral normalisation was proposed to address mode collapse and improve training stability. The experiment shows that CSR-GAN has better performance in reconstructing the image with higher frequency details and reducing noise compared to other methods, with an optimised PSNR of 34.6 and SSIM of 0.89. However, SOUP-GAN performed the best in terms of delivering less noisy images with good structures, achieving a PSNR of 34.4 and SSIM of 0.83. The obtained results indicate that the proposed enhanced GAN model can be a useful tool for MR image quality improvement for subsequent better disease diagnostics.
Authors: Jiayang Shi, Lincen Yang, Zhong Li, Tristan Van Leeuwen, Daniel M. Pelt, K. Joost Batenburg
Abstract: Generative models, particularly Diffusion Models (DM), have shown strong potential for Computed Tomography (CT) reconstruction serving as expressive priors for solving ill-posed inverse problems. However, diffusion-based reconstruction relies on Stochastic Differential Equations (SDEs) for forward diffusion and reverse denoising, where such stochasticity can interfere with repeated data consistency corrections in CT reconstruction. Since CT reconstruction is often time-critical in clinical and interventional scenarios, improving reconstruction efficiency is essential. In contrast, Flow Matching (FM) models sampling as a deterministic Ordinary Differential Equation (ODE), yielding smooth trajectories without stochastic noise injection. This deterministic formulation is naturally compatible with repeated data consistency operations. Furthermore, we observe that FM-predicted velocity fields exhibit strong correlations across adjacent steps. Motivated by this, we propose an FM-based CT reconstruction framework (FMCT) and an efficient variant (EFMCT) that reuses previously predicted velocity fields over consecutive steps to substantially reduce the number of Neural network Function Evaluations (NFEs), thereby improving inference efficiency. We provide theoretical analysis showing that the error introduced by velocity reuse is bounded when combined with data consistency operations. Extensive experiments demonstrate that FMCT/EFMCT achieve competitive reconstruction quality while significantly improving computational efficiency compared with diffusion-based methods. The codebase is open-sourced at https://github.com/EFMCT/EFMCT.
Authors: Yunzheng Zhu, Aichi Chien, Kimaya kulkarni, Luoting Zhuang, Stephen Park, Ricky Savjani, Daniel Low, William Hsu
Abstract: Deformable registration is crucial in medical imaging. Several existing applications include lesion tracking, probabilistic atlas generation, and treatment response evaluation. However, current methods often lack robustness and generalizability across two key factors: spatial resolution and differences in anatomical coverage. We jointly optimize a registration field and a learnable dimensionality reduction module so that compressed VFM embeddings remain registration-relevant, and fuse these global semantic cues with MIND local descriptors. GLIDE-Reg achieves average dice similarity coefficients (DSC) across 6 anatomical structures of 0.859, 0.862, and 0.901 in two public cohorts (Lung250M and NLST) and one institution cohort (UCLA5DCT), and outperforms the state-of-the-art DEEDS (0.834, 0.858, 0.900) with relative improvements of 3.0%, 0.5%, and 0.1%. For target registration errors, GLIDE-Reg achieves 1.58 mm on Lung250M landmarks (compared to 1.25 mm on corrField and 1.91 mm on DEEDS) and 1.11 mm on NLST nodule centers (compared to 1.11 mm on DEEDS). The substantiated performance on the nodule centers also demonstrates its robustness across challenging downstream tasks, such as nodule tracking, which is an essential prior step for early-stage lung cancer diagnosis.
Authors: Jonas J\"ager, Florian J. Kiwit, Carlos A. Riofr\'io
Abstract: Quantum generative modeling is a rapidly evolving discipline at the intersection of quantum computing and machine learning. Contemporary quantum machine learning is generally limited to toy examples or heavily restricted datasets with few elements. This is not only due to the current limitations of available quantum hardware but also due to the absence of inductive biases arising from application-agnostic designs. Current quantum solutions must resort to tricks to scale down high-resolution images, such as relying heavily on dimensionality reduction or utilizing multiple quantum models for low-resolution image patches. Building on recent developments in classical image loading to quantum computers, we circumvent these limitations and train quantum Wasserstein GANs on the established classical MNIST and Fashion-MNIST datasets. Using the complete datasets, our system generates full-resolution images across all ten classes and establishes a new state-of-the-art performance with a single end-to-end quantum generator without tricks. As a proof-of-principle, we also demonstrate that our approach can be extended to color images, exemplified on the Street View House Numbers dataset. We analyze how the choice of variational circuit architecture introduces inductive biases, which crucially unlock this performance. Furthermore, enhanced noise input techniques enable highly diverse image generation while maintaining quality. Finally, we show promising results even under quantum shot noise conditions.
Authors: Hutama Arif Bramantyo, Mukarram Ali Faridi, Rui Chen, Clarissa Harris, Yin Sun
Abstract: In this study, we present a meat freshness classification framework from Red-Green-Blue (RGB) images that supports both packaged and unpackaged meat datasets. The system classifies four in-distribution (ID) meat classes and uses an out-of-distribution (OOD)-aware abstention mechanism that flags low-confidence samples as No Result. The pipeline combines U-Net-based segmentation with deep feature classifiers. Segmentation is used as a preprocessing step to isolate the meat region and reduce background, producing more consistent inputs for classification. The segmentation module achieved an Intersection over Union (IoU) of 75% and a Dice coefficient of 82%, producing standardized inputs for the classification stage. For classification, we benchmark five backbones: Residual Network-50 (ResNet-50), Vision Transformer-Base/16 (ViT-B/16), Swin Transformer-Tiny (Swin-T), EfficientNet-B0, and MobileNetV3-Small. We use nested 5x3 cross-validation (CV) for model selection and hyperparameter tuning. On the held-out ID test set, EfficientNet-B0 achieves the highest accuracy (98.10%), followed by ResNet-50 and MobileNetV3-Small (both 97.63%) and Swin-T (97.51%), while ViT-B/16 is lower (94.42%). We additionally evaluate OOD scoring and thresholding using standard OOD metrics and sensitivity analysis over the abstention threshold. Finally, we report on-device latency using TensorFlow Lite (TFLite) on a smartphone, highlighting practical accuracy-latency trade-offs for future deployment.
Authors: Cunyuan Yang, Dejuan Song, Xiaotao Pang, Qianqian Shen, Wenjie Nie, Yifan Huang, Lei Wu, Wei Han, Haishuai Wang, Jiajun Bu
Abstract: The automatic generation of medical reports utilizing Multimodal Large Language Models (MLLMs) frequently encounters challenges related to factual instability, which may manifest as the omission of findings or the incorporation of inaccurate information, thereby constraining their applicability in clinical settings. Current methodologies typically produce reports based directly on image features, which inherently lack a definitive factual basis. In response to this limitation, we introduce Fact-Flow, an innovative framework that separates the process of visual fact identification from the generation of reports. This is achieved by initially predicting clinical findings from the image, which subsequently directs the MLLM to produce a report that is factually precise. A pivotal advancement of our approach is a pipeline that leverages a Large Language Model (LLM) to autonomously create a dataset of labeled medical findings, effectively eliminating the need for expensive manual annotation. Extensive experimental evaluations conducted on two disease-focused medical datasets validate the efficacy of our method, demonstrating a significant enhancement in factual accuracy compared to state-of-the-art models, while concurrently preserving high standards of text quality.
Authors: Xu Luo, Ji Zhang, Lianli Gao, Heng Tao Shen, Jingkuan Song
Abstract: Few-shot transfer has been revolutionized by stronger pre-trained models and improved adaptation algorithms.However, there lacks a unified, rigorous evaluation protocol that is both challenging and realistic for real-world usage. In this work, we establish FEWTRANS, a comprehensive benchmark containing 10 diverse datasets, and propose the Hyperparameter Ensemble (HPE) protocol to overcome the "validation set illusion" in data-scarce regimes. Our empirical findings demonstrate that the choice of pre-trained model is the dominant factor for performance, while many sophisticated transfer methods offer negligible practical advantages over a simple full-parameter fine-tuning baseline. To explain this surprising effectiveness, we provide an in-depth mechanistic analysis showing that full fine-tuning succeeds via distributed micro-adjustments and more flexible reshaping of high-level semantic presentations without suffering from overfitting. Additionally, we quantify the performance collapse of multimodal models in specialized domains as a result of linguistic rarity using adjusted Zipf frequency scores. By releasing FEWTRANS, we aim to provide a rigorous "ruler" to streamline reproducible advances in few-shot transfer learning research. We make the FEWTRANS benchmark publicly available at https://github.com/Frankluox/FewTrans.
Authors: Khaleque Md Aashiq Kamal, Surya Eada, Aayushi Verma, Subek Acharya, Adrian Yemin, Benjamin Fuller, Kaleel Mahmood
Abstract: Developments in the machine learning voting domain have shown both promising results and risks. Trained models perform well on ballot classification tasks (> 99% accuracy) but are at risk from adversarial example attacks that cause misclassifications. In this paper, we analyze an attacker who seeks to deploy adversarial examples against machine learning ballot classifiers to compromise a U.S. election. We first derive a probabilistic framework for determining the number of adversarial example ballots that must be printed to flip an election, in terms of the probability of each candidate winning and the total number of ballots cast. Second, it is an open question as to which type of adversarial example is most effective when physically printed in the voting domain. We analyze six different types of adversarial example attacks: l_infinity-APGD, l2-APGD, l1-APGD, l0 PGD, l0 + l_infinity PGD, and l0 + sigma-map PGD. Our experiments include physical realizations of 144,000 adversarial examples through printing and scanning with four different machine learning models. We empirically demonstrate an analysis gap between the physical and digital domains, wherein attacks most effective in the digital domain (l2 and l_infinity) differ from those most effective in the physical domain (l1 and l2, depending on the model). By unifying a probabilistic election framework with digital and physical adversarial example evaluations, we move beyond prior close race analyses to explicitly quantify when and how adversarial ballot manipulation could alter outcomes.
Authors: Zeyu Chen, Huanjin Yao, Ziwang Zhao, Min Yang
Abstract: Using Multimodal Large Language Models (MLLMs) as judges to achieve precise and consistent evaluations has gradually become an emerging paradigm across various domains. Evaluating the capability and reliability of MLLM-as-a-judge systems is therefore essential for ensuring trustworthy assessment. Existing judge benchmarks categorize samples by task types but fail to capture the fundamental judgment capabilities required for reliable evaluation. In this work, we introduce M-JudgeBench, a ten-dimensional capability-oriented benchmark designed to comprehensively assess the judgment abilities of MLLMs. Our benchmark decomposes evaluation into pairwise Chain-of-Thought (CoT) comparison, length bias avoidance, and process error detection tasks, jointly covering ten fine-grained subtasks. This design enables diagnosis of model reliability across reasoning styles, response lengths, and cross-model variations. Systematic evaluation uncovers the systematic weaknesses in existing MLLM-as-a-judge systems. To address this issue, we further propose Judge-MCTS, a data construction framework generating pairwise reasoning trajectories with various correctness and length. Using Judge-MCTS, we construct an MCTS-augmented dataset and train M-Judger, a series of strong judge models. Extensive experiments demonstrate the superiority of M-Judger on existing judge benchmarks as well as M-JudgeBench. Overall, our work establishes a more principled foundation for evaluating MLLM-as-a-judge through M-JudgeBench and Judge-MCTS framework, paving the way for future research on judge model evaluation and capability-driven judge training.
Authors: Rongsheng Wang, Minghao Wu, Hongru Zhou, Zhihan Yu, Zhenyang Cai, Junying Chen, Benyou Wang
Abstract: Recent advances in video generation have opened new avenues for macroscopic simulation of complex dynamic systems, but their application to microscopic phenomena remains largely unexplored. Microscale simulation holds great promise for biomedical applications such as drug discovery, organ-on-chip systems, and disease mechanism studies, while also showing potential in education and interactive visualization. In this work, we introduce MicroWorldBench, a multi-level rubric-based benchmark for microscale simulation tasks. MicroWorldBench enables systematic, rubric-based evaluation through 459 unique expert-annotated criteria spanning multiple microscale simulation task (e.g., organ-level processes, cellular dynamics, and subcellular molecular interactions) and evaluation dimensions (e.g., scientific fidelity, visual quality, instruction following). MicroWorldBench reveals that current SOTA video generation models fail in microscale simulation, showing violations of physical laws, temporal inconsistency, and misalignment with expert criteria. To address these limitations, we construct MicroSim-10K, a high-quality, expert-verified simulation dataset. Leveraging this dataset, we train MicroVerse, a video generation model tailored for microscale simulation. MicroVerse can accurately reproduce complex microscale mechanism. Our work first introduce the concept of Micro-World Simulation and present a proof of concept, paving the way for applications in biology, education, and scientific visualization. Our work demonstrates the potential of educational microscale simulations of biological mechanisms. Our data and code are publicly available at https://github.com/FreedomIntelligence/MicroVerse
Authors: Yuchen Hou, Lin Zhao
Abstract: Vision-Language-Action (VLA) models achieve over 95% success on standard benchmarks. However, through systematic experiments, we find that current state-of-the-art VLA models largely ignore language instructions. Prior work lacks: (1) systematic semantic perturbation diagnostics, (2) a benchmark that forces language understanding by design, and (3) linguistically diverse training data. This paper constructs the LangGap benchmark, based on a four-dimensional semantic perturbation method -- varying instruction semantics while keeping the tabletop layout fixed -- revealing language understanding deficits in {\pi}0.5. Existing benchmarks like LIBERO assign only one task per layout, underutilizing available objects and target locations; LangGap fully diversifies pick-and-place tasks under identical layouts, forcing models to truly understand language. Experiments show that targeted data augmentation can partially close the language gap -- success rate improves from 0% to 90% with single-task training, and 0% to 28% with multi-task training. However, as semantic diversity of extended tasks increases, model learning capacity proves severely insufficient; even trained tasks perform poorly. This reveals a fundamental challenge for VLA models in understanding diverse language instructions -- precisely the long-term value of LangGap.
Authors: Zhanwang Liu, Yuting Li, Haoyuan Gao, Yexin Li, Linghe Kong, Lichao Sun, Weiran Huang
Abstract: Catastrophic forgetting, the tendency of neural networks to forget previously learned knowledge when learning new tasks, has been a major challenge in continual learning (CL). To tackle this challenge, CL methods have been proposed and shown to reduce forgetting. Furthermore, CL models deployed in mission-critical settings can benefit from uncertainty awareness by calibrating their predictions to reliably assess their confidences. However, existing uncertainty-aware continual learning methods suffer from high computational overhead and incompatibility with mainstream replay methods. To address this, we propose idempotent experience replay (IDER), a novel approach based on the idempotent property where repeated function applications yield the same output. Specifically, we first adapt the training loss to make model idempotent on current data streams. In addition, we introduce an idempotence distillation loss. We feed the output of the current model back into the old checkpoint and then minimize the distance between this reprocessed output and the original output of the current model. This yields a simple and effective new baseline for building reliable continual learners, which can be seamlessly integrated with other CL approaches. Extensive experiments on different CL benchmarks demonstrate that IDER consistently improves prediction reliability while simultaneously boosting accuracy and reducing forgetting. Our results suggest the potential of idempotence as a promising principle for deploying efficient and trustworthy continual learning systems in real-world applications.Our code is available at https://github.com/YutingLi0606/Idempotent-Continual-Learning.
URLs: https://github.com/YutingLi0606/Idempotent-Continual-Learning.
Authors: Hsin Lin, Yan-Lun Chen, Ren-Hung Hwang, Chia-Mu Yu
Abstract: Backdoor attacks pose a critical threat to the security of deep neural networks, yet existing efforts on universal backdoors often rely on visually salient patterns, making them easier to detect and less practical at scale. In this work, we introduce a novel imperceptible universal backdoor attack that simultaneously controls all target classes with minimal poisoning while preserving stealth. Our key idea is to leverage graph convolutional networks (GCNs) to model inter-class relationships and generate class-specific perturbations that are both effective and visually invisible. The proposed framework optimizes a dual-objective loss that balances stealthiness (measured by perceptual similarity metrics such as PSNR) and attack success rate (ASR), enabling scalable, multi-target backdoor injection. Extensive experiments on ImageNet-1K with ResNet architectures demonstrate that our method achieves high ASR (up to 91.3%) under poisoning rates as low as 0.16%, while maintaining benign accuracy and evading state-of-the-art defenses. These results highlight the emerging risks of invisible universal backdoors and call for more robust detection and mitigation strategies.
Authors: Zhenhao Zhang, Jiaxin Liu, Ye Shi, Jingya Wang
Abstract: Planning physically feasible dexterous hand manipulation is a central challenge in robotic manipulation and Embodied AI. Prior work typically relies on object-centric cues or precise hand-object interaction sequences, foregoing the rich, compositional guidance of open-vocabulary instruction. We introduce UniHM, the first framework for unified dexterous hand manipulation guided by free-form language commands. We propose a Unified Hand-Dexterous Tokenizer that maps heterogeneous dexterous-hand morphologies into a single shared codebook, improving cross-dexterous hand generalization and scalability to new morphologies. Our vision language action model is trained solely on human-object interaction data, eliminating the need for massive real-world teleoperation datasets, and demonstrates strong generalizability in producing human-like manipulation sequences from open-ended language instructions. To ensure physical realism, we introduce a physics-guided dynamic refinement module that performs segment-wise joint optimization under generative and temporal priors, yielding smooth and physically feasible manipulation sequences. Across multiple datasets and real-world evaluations, UniHM attains state-of-the-art results on both seen and unseen objects and trajectories, demonstrating strong generalization and high physical feasibility. Our project page at \href{https://unihm.github.io/}{https://unihm.github.io/}.
Authors: Matt Y. Cheung, Ashok Veeraraghavan, Guha Balakrishnan
Abstract: Template-based segmentation, a widely used paradigm in medical imaging, propagates anatomical labels via deformable registration from a labeled atlas to a target image, and is often used to compute volumetric biomarkers for downstream decision-making. While conformal prediction (CP) provides finite-sample valid intervals for scalar metrics, existing segmentation-based uncertainty quantification (UQ) approaches either rely on learned model features, often unavailable in classic template-based pipelines, or treat the registration process as a black box, resulting in overly conservative intervals when applied directly in output space. We introduce ConVOLT, a CP framework that achieves efficient volumetric UQ by conditioning calibration on properties of the estimated deformation field from template-based segmentation. ConVOLT calibrates a learned volumetric scaling factor from deformation space features. We evaluate ConVOLT on template-based segmentation tasks involving global, regional, and label volumetry across multiple datasets and registration methods. ConVOLT achieves target coverage while producing substantially tighter intervals than output-space conformal baselines. Our work paves way to exploit the registration process for efficient UQ in medical imaging pipelines.
Authors: Zhangxing Bian, Shuwen Wei, Samuel W. Remedios, Junyu Chen, Aaron Carass, Blake E. Dewey, Jerry L. Prince
Abstract: Tagged MRI enables tracking internal tissue motion non-invasively. It encodes motion by modulating anatomy with periodic tags, which deform along with tissue. However, the entanglement between anatomy, tags and motion poses significant challenges for post-processing. The existence of tags and imaging blur hinders downstream tasks such as segmenting anatomy. Tag fading, due to T1-relaxation, disrupts the brightness constancy assumption for motion tracking. For decades, these challenges have been handled in isolation and sub-optimally. In contrast, we introduce a blind and nonlinear inverse framework for tagged MRI that, for the first time, unifies these tasks: anatomical image recovery, high-resolution cine image synthesis, and motion estimation. At its core, the synergy of MR physics and generative priors enables us to blindly estimate the unknown forward imaging models and high-resolution underlying anatomy, while simultaneously tracking 3D diffeomorphic Lagrangian motion over time. Experiments on tagged brain MRI demonstrate that our approach yields high-resolution anatomy images, cine images, and more accurate motion than specialized methods.
Authors: Li Lucy, Albert Zhang, Nathan Anderson, Ryan Knight, Kyle Lo
Abstract: Effective mathematics education requires identifying and responding to students' mistakes. For AI to support pedagogical applications, models must perform well across different levels of student proficiency. Our work provides an extensive, year-long snapshot of how 11 vision-language models (VLMs) perform on DrawEduMath, a QA benchmark involving real students' handwritten, hand-drawn responses to math problems. We find that models' weaknesses concentrate on a core component of math education: student error. All evaluated VLMs underperform when describing work from students who require more pedagogical help, and across all QA, they struggle the most on questions related to assessing student error. Thus, while VLMs may be optimized to be math problem solving experts, our results suggest that they require alternative development incentives to adequately support educational use cases.
Authors: Joshua Steier
Abstract: Contrastive Forward-Forward (CFF) learning trains Vision Transformers layer by layer against supervised contrastive objectives. CFF training can be sensitive to random seed, but the sources of this instability are poorly understood. We focus on one implementation detail: the positive-pair margin in the contrastive loss is applied through saturating similarity clamping, $\min(s + m,\, 1)$. We prove that an alternative formulation, subtracting the margin after the log-probability, is gradient-neutral under the mean-over-positives reduction. On CIFAR-10 ($2 \times 2$ factorial, $n{=}7$ seeds per cell), clamping produces $5.90\times$ higher pooled test-accuracy variance ($p{=}0.003$) with no difference in mean accuracy. Analyses of clamp activation rates, layerwise gradient norms, and a reduced-margin probe point to saturation-driven gradient truncation at early layers. The effect does not transfer cleanly to other datasets: on CIFAR-100, SVHN, and Fashion-MNIST, clamping produces equal or lower variance. Two factors account for the discrepancy. First, positive-pair density per batch controls how often saturation occurs. Second, task difficulty compresses seed-to-seed spread when accuracy is high. An SVHN difficulty sweep confirms the interaction on a single dataset, with the variance ratio moving from $0.25\times$ at high accuracy to $16.73\times$ under aggressive augmentation. In moderate-accuracy regimes with many same-class pairs per batch, switching to the gradient-neutral subtraction reference removes this variance inflation at no cost to mean accuracy. Measuring the layer-0 clamp activation rate serves as a simple check for whether the problem applies.
Authors: Susmita Ghanta, Karan Nathwani, Rohit Chaurasiya
Abstract: Real-time unmanned aerial vehicle (UAV) acoustic detection at the edge demands low-latency inference under strict power and hardware limits. This paper presents SHIELD8-UAV, a sequential 8-bit hardware implementation of a precision-aware 1D feature-driven CNN (1D-F-CNN) accelerator for continuous acoustic monitoring. The design performs layer-wise execution on a shared multi-precision datapath, eliminating the need for replicated processing elements. A layer-sensitivity quantisation framework supports FP32, BF16, INT8, and FXP8 modes, while structured channel pruning reduces the flattened feature dimension from 35,072 to 8,704 (75%), thereby lowering serialised dense-layer cycles. The model achieves 89.91% detection accuracy in FP32 with less than 2.5% degradation in 8-bit modes. The accelerator uses 2,268 LUTs and 0.94 W power with 116 ms end-to-end latency, achieving 37.8% and 49.6% latency reduction compared with QuantMAC and LPRE, respectively, on a Pynq-Z2 FPGA, and 5-9% lower logic usage than parallel designs. ASIC synthesis in UMC 40 nm technology shows a maximum operating frequency of 1.56 GHz, 3.29 mm2 core area, and 1.65 W total power. These results demonstrate that sequential execution combined with precision-aware quantisation and serialisation-aware pruning enables practical low-energy edge inference without relying on massive parallelism.
Authors: Sicheng Yang, Yukai Huang, Weitong Cai, Shitong Sun, Fengyi Fang, You He, Yiqiao Xie, Jiankang Deng, Hang Zhang, Jifei Song, Zhensong Zhang
Abstract: What if accessing the web did not require a screen, a stable desk, or even free hands? For people navigating crowded cities, living with low vision, or experiencing cognitive overload, smart glasses coupled with AI agents could turn the web into an always-on assistive layer over daily life. We present Egocentric Co-Pilot, a web-native neuro-symbolic framework that runs on smart glasses and uses a Large Language Model (LLM) to orchestrate a toolbox of perception, reasoning, and web tools. An egocentric reasoning core combines Temporal Chain-of-Thought with Hierarchical Context Compression to support long-horizon question answering and decision support over continuous first-person video, far beyond a single model's context window. Additionally, a lightweight multimodal intent layer maps noisy speech and gaze into structured commands. We further implement and evaluate a cloud-native WebRTC pipeline integrating streaming speech, video, and control messages into a unified channel for smart glasses and browsers. In parallel, we deploy an on-premise WebSocket baseline, exposing concrete trade-offs between local inference and cloud offloading in terms of latency, mobility, and resource use. Experiments on Egolife and HD-EPIC demonstrate competitive or state-of-the-art egocentric QA performance, and a human-in-the-loop study on smart glasses shows higher task completion and user satisfaction than leading commercial baselines. Taken together, these results indicate that web-connected egocentric co-pilots can be a practical path toward more accessible, context-aware assistance in everyday life. By grounding operation in web-native communication primitives and modular, auditable tool use, Egocentric Co-Pilot offers a concrete blueprint for assistive, always-on web agents that support education, accessibility, and social inclusion for people who may benefit most from contextual, egocentric AI.
Authors: Haozhe Lou, Mingtong Zhang, Haoran Geng, Hanyang Zhou, Sicheng He, Zhiyuan Gao, Siheng Zhao, Jiageng Mao, Pieter Abbeel, Jitendra Malik, Daniel Seita, Yue Wang
Abstract: Simulation provides a cost-effective and flexible platform for data generation and policy learning to develop robotic systems. However, bridging the gap between simulation and real-world dynamics remains a significant challenge, especially in physical parameter identification. In this work, we introduce a real-to-sim-to-real engine that leverages the Gaussian Splat representations to build a differentiable engine, enabling object mass identification from real-world visual observations and robot control signals, while enabling grasping policy learning simultaneously. Through optimizing the mass of the manipulated object, our method automatically builds high-fidelity and physically plausible digital twins. Additionally, we propose a novel approach to train force-aware grasping policies from limited data by transferring feasible human demonstrations into simulated robot demonstrations. Through comprehensive experiments, we demonstrate that our engine achieves accurate and robust performance in mass identification across various object geometries and mass values. Those optimized mass values facilitate force-aware policy learning, achieving superior and high performance in object grasping, effectively reducing the sim-to-real gap.
Authors: Zhexin Xu, Nikolas R. Sanderson, Hanna Jiamei Zhang, David M. Rosen
Abstract: Factor graphs provide a convenient modular modeling language that enables practitioners to design and deploy high-performance robotic state estimation systems by composing simple, reusable building blocks. However, inference in these models is typically performed using local optimization methods that can converge to suboptimal solutions, a serious reliability concern in safety-critical applications. Conversely, certifiable estimators based on convex relaxation can recover verifiably globally optimal solutions in many practical settings, but the computational cost of solving their large-scale relaxations necessitates specialized, structure-exploiting solvers that require substantial expertise to implement, significantly hampering practical deployment. In this paper, we show that these two paradigms, which have thus far been treated as independent in the literature, can be naturally synthesized into a unified framework for certifiable factor graph optimization. The key insight is that factor graph structure is preserved under Shor's relaxation and Burer-Monteiro factorization: applying these transformations to a QCQP with an associated factor graph representation yields a lifted problem admitting a factor graph model with identical connectivity, in which variables and factors are simple one-to-one algebraic transformations of those in the original QCQP. This structural preservation enables the Riemannian Staircase methodology for certifiable estimation to be implemented using the same mature, highly-performant factor graph libraries and workflows already ubiquitously employed throughout robotics and computer vision, making certifiable estimation as straightforward to design and deploy as conventional factor graph inference.
Authors: Xiangjian Hou, Chao Qin, Chang Ni, Xin Wang, Chun Yuan, Xiaodong Ma
Abstract: Global token mixing, implemented via self-attention or state-space sequence models, has become a popular model design choice for MRI restoration. However, MRI restoration tasks differ substantially in how their degradations vary over image and k-space domains, and in the degree to which global coupling is already imposed by physics-driven data consistency terms. In this work, we ask the question whether global token mixing is actually beneficial in each individual task across three representative settings: accelerated MRI reconstruction with explicit data consistency, MRI super-resolution with k-space center cropping, and denoising of clinical carotid MRI data with spatially heteroscedastic noise. To reduce confounding factors, we establish a controlled testbed comparing a minimal local gated CNN and its large-field variant, benchmarking them directly against state-of-the-art global models under aligned training and evaluation protocols. For accelerated MRI reconstruction, the minimal unrolled gated-CNN baseline is already highly competitive compared to recent token-mixing approaches in public reconstruction benchmarks, suggesting limited additional benefits when the forward model and data-consistency steps provide strong global constraints. For super-resolution, where low-frequency k-space data are largely preserved by the controlled low-pass degradation, local gated models remain competitive, and a lightweight large-field variant yields only modest improvements. In contrast, for denoising with pronounced spatially heteroscedastic noise, token-mixing models achieve the strongest overall performance, consistent with the need to estimate spatially varying reliability. In conclusion, our results demonstrate that the utility of global token mixing in MRI restoration is task-dependent, and it should be tailored to the underlying imaging physics and degradation structure.
Authors: Tianyi Xu, Rong Shan, Junjie Wu, Jiadeng Huang, Teng Wang, Jiachen Zhu, Wenteng Chen, Minxin Tu, Quantao Dou, Zhaoxiang Wang, Changwang Zhang, Weinan Zhang, Jun Wang, Jianghao Lin
Abstract: Personal photo albums are not merely collections of static images but living, ecological archives defined by temporal continuity, social entanglement, and rich metadata, which makes the personalized photo retrieval non-trivial. However, existing retrieval benchmarks rely heavily on context-isolated web snapshots, failing to capture the multi-source reasoning required to resolve authentic, intent-driven user queries. To bridge this gap, we introduce PhotoBench, the first benchmark constructed from authentic, personal albums. It is designed to shift the paradigm from visual matching to personalized multi-source intent-driven reasoning. Based on a rigorous multi-source profiling framework, which integrates visual semantics, spatial-temporal metadata, social identity, and temporal events for each image, we synthesize complex intent-driven queries rooted in users' life trajectories. Extensive evaluation on PhotoBench exposes two critical limitations: the modality gap, where unified embedding models collapse on non-visual constraints, and the source fusion paradox, where agentic systems perform poor tool orchestration. These findings indicate that the next frontier in personal multimodal retrieval lies beyond unified embeddings, necessitating robust agentic reasoning systems capable of precise constraint satisfaction and multi-source fusion. Our PhotoBench is available.
Authors: Leyla Roksan Caglar, Pedro A. M. Mediano, Baihan Lin
Abstract: Generalization to novel visual conditions remains a central challenge for both human and machine vision, yet standard robustness metrics offer limited insight into how systems trade accuracy for robustness. We introduce a rate-distortion-theoretic framework that treats stimulus-response behavior as an effective communication channel, derives rate-distortion (RD) frontiers from confusion matrices, and summarizes each system with two interpretable geometric signatures - slope ($\beta$) and curvature ($\kappa$) - which capture the marginal cost and abruptness of accuracy-robustness trade-offs. Applying this framework to human psychophysics and 18 deep vision models under controlled image perturbations, we compare generalization geometry across model architectures and training regimes. We find that both biological and artificial systems follow a common lossy-compression principle but occupy systematically different regions of RD space. In particular, humans exhibit smoother, more flexible trade-offs, whereas modern deep networks operate in steeper and more brittle regimes even at matched accuracy. Across training regimes, robustness training induces systematic but dissociable shifts in beta/kappa, revealing cases where improved robustness or accuracy does not translate into more human-like generalization geometry. These results demonstrate that RD geometry provides a compact, model-agnostic lens for comparing generalization behavior across systems beyond standard accuracy-based metrics.
Authors: Minwoo Kim, Seunghyeok Shin, Hongki Lim
Abstract: Training-free diffusion priors enable inverse-problem solvers without retraining, but for nonlinear forward operators data consistency often relies on repeated derivatives or inner optimization/MCMC loops with conservative step sizes, incurring many iterations and denoiser/score evaluations. We propose a training-free solver that replaces these inner loops with a hard measurement-space feasibility constraint (closed-form projection) and an analytic, model-optimal step size, enabling a small, fixed compute budget per noise level. Anchored at the denoiser prediction, the correction is approximated via an adjoint-free, ADMM-style splitting with projection and a few steepest-descent updates, using one VJP and either one JVP or a forward-difference probe, followed by backtracking and decoupled re-annealing. We prove local model optimality and descent under backtracking for the step-size rule, and derive an explicit KL bound for mode-substitution re-annealing under a local Gaussian conditional surrogate. We also develop a latent variant and a one-parameter pixel$\rightarrow$latent hybrid schedule. Experiments achieve competitive PSNR/SSIM/LPIPS with up to 19.5$\times$ speedup, without hand-coded adjoints or inner MCMC.
Authors: Xiaohao Liu, Xiaobo Xia, Manyi Zhang, Ji-Fu Li, Xianzhi Yu, Fei Shen, Xiu Su, See-Kiong Ng, Tat-Seng Chua
Abstract: Quantization is pivotal for mitigating the significant memory and computational overhead of Large Language Models (LLMs). While emerging transformation-based methods have successfully enhanced quantization by projecting feature spaces onto smoother manifolds using orthogonal matrices, they typically enforce a rigid one-to-one transformation constraint. This static approach fails to account for the dynamic patterns inherent in input activations, particularly within diffusion LLMs (dLLMs) and Multimodal LLMs (MLLMs), where varying token types exhibit distinct distributions. To advance this, we propose FreeAct, a novel quantization framework that relaxes the static one-to-one constraint to accommodate dynamic activation disparities. Theoretically, we leverage the rank-deficient nature of activations to derive a solution space that extends beyond simple inverse matrices, enabling the decoupling of activation transformations from weights. Methodologically, FreeAct identifies token-specific dynamics (i.e., vision v.s. text, or masked tokens) and allocates distinct transformation matrices to the activation side, while maintaining a unified, static transformation for the weights. Extensive experiments across dLLMs and MLLMs demonstrate that FreeAct significantly outperforms baselines, up to 5.3% performance improvement, with in-depth analyses. Our code will be publicly released.
Authors: Ilenia Carboni, Elia Cereda, Lorenzo Lamberti, Daniele Malpetti, Francesco Conti, Daniele Palossi
Abstract: Sub-30g nano-sized aerial robots can leverage their agility and form factor to autonomously explore cluttered and narrow environments, like in industrial inspection and search and rescue missions. However, the price for their tiny size is a strong limit in their resources, i.e., sub-100 mW microcontroller units (MCUs) delivering $\sim$100 GOps/s at best, and memory budgets well below 100 MB. Despite these strict constraints, we aim to enable complex vision-based tasks aboard nano-drones, such as dense 3D scene reconstruction: a key robotic task underlying fundamental capabilities like spatial awareness and motion planning. Top-performing 3D reconstruction methods leverage neural radiance fields (NeRF) models, which require GBs of memory and massive computation, usually delivered by high-end GPUs consuming 100s of Watts. Our work introduces Tiny-DroNeRF, a lightweight NeRF model, based on Instant-NGP, and optimized for running on a GAP9 ultra-low-power (ULP) MCU aboard our nano-drones. Then, we further empower our Tiny-DroNeRF by leveraging a collaborative federated learning scheme, which distributes the model training among multiple nano-drones. Our experimental results show a 96% reduction in Tiny-DroNeRF's memory footprint compared to Instant-NGP, with only a 5.7 dB drop in reconstruction accuracy. Finally, our federated learning scheme allows Tiny-DroNeRF to train with an amount of data otherwise impossible to keep in a single drone's memory, increasing the overall reconstruction accuracy. Ultimately, our work combines, for the first time, NeRF training on an ULP MCU with federated learning on nano-drones.
Authors: Xinxin Dong, Haokai Ma, Yuze Zheng, Yongfu Zha, Yonghui Yang, Xiaodong Wang
Abstract: Micro-video recommendation aims to capture user preferences from the collaborative and context information of the interacted micro-videos, thereby predicting the appropriate videos. This target is often hindered by the inherent noise within multimodal content and unreliable implicit feedback, which weakens the correspondence between behaviors and underlying interests. While conventional works have predominantly approached such scenario through behavior-augmented modeling and content-centric multimodal analysis, these paradigms can inadvertently give rise to two non-trivial challenges: preference-irrelative video representation extraction and inherent modality conflicts. To address these issues, we propose a Multi-granularity sequential modeling method via hierarchical diffusion models for micro-video Recommendation (MealRec), which simultaneously considers temporal correlations during preference modeling from intra- and inter-video perspectives. Specifically, we first propose Temporal-guided Content Diffusion (TCD) to refine video representations under intra-video temporal guidance and personalized collaborative signals to emphasize salient content while suppressing redundancy. To achieve the semantically coherent preference modeling, we further design the Noise-unconditional Preference Denoising (NPD) to recovers informative user preferences from corrupted states under the blind denoising. Extensive experiments and analyses on four micro-video datasets from two platforms demonstrate the effectiveness, universality, and robustness of our MealRec, further uncovering the effective mechanism of our proposed TCD and NPD. The source code and corresponding dataset will be available upon acceptance.
Authors: Christopher Driggers-Ellis, Nachiketh Tibrewal, Rohit Bogulla, Harsh Khanna, Sangpil Youm, Christan Grant, Bonnie Dorr
Abstract: A system that enables blind or visually impaired users to access comics/manga would introduce a new medium of storytelling to this community. However, no such system currently exists. Generative vision-language models (VLMs) have shown promise in describing images and understanding comics, but most research on comic understanding is limited to panel-level analysis. To fully support blind and visually impaired users, greater attention must be paid to page-level understanding and interpretation. In this work, we present a preliminary benchmark of VLM performance on comic interpretation tasks. We identify and categorize hallucinations that emerge during this process, organizing them into generalized object-hallucination taxonomies. We conclude with guidance on future research, emphasizing hallucination mitigation and improved data curation for comic interpretation.
Authors: Pengyuan Wu, Pingrui Zhang, Zhigang Wang, Dong Wang, Bin Zhao, Xuelong Li
Abstract: Diffusion-based policies have achieved remarkable results in robotic manipulation but often struggle to adapt rapidly in dynamic scenarios, leading to delayed responses or task failures. We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction. DCDP integrates a self-supervised dynamic feature encoder, cross-attention fusion, and an asymmetric action encoder-decoder to inject environmental dynamics before action execution, achieving real-time closed-loop action correction and enhancing the system's adaptability in dynamic scenarios. In dynamic PushT simulations, DCDP improves adaptability by 19\% without retraining while requiring only 5\% additional computation. Its modular design enables plug-and-play integration, achieving both temporal coherence and real-time responsiveness in dynamic robotic scenarios, including real-world manipulation tasks. The project page is at: https://github.com/wupengyuan/dcdp
Authors: Jingbiao Mei, Jinghong Chen, Guangyu Yang, Xinyu Hou, Margaret Li, Bill Byrne
Abstract: Personalized AI assistants must recall and reason over long-term user memory, which naturally spans multiple modalities and sources such as images, videos, and emails. However, existing Long-term Memory benchmarks focus primarily on dialogue history, failing to capture realistic personalized references grounded in lived experience. We introduce ATM-Bench, the first benchmark for multimodal, multi-source personalized referential Memory QA. ATM-Bench contains approximately four years of privacy-preserving personal memory data and human-annotated question-answer pairs with ground-truth memory evidence, including queries that require resolving personal references, multi-evidence reasoning from multi-source and handling conflicting evidence. We propose Schema-Guided Memory (SGM) to structurally represent memory items originated from different sources. In experiments, we implement 5 state-of-the-art memory systems along with a standard RAG baseline and evaluate variants with different memory ingestion, retrieval, and answer generation techniques. We find poor performance (under 20\% accuracy) on the ATM-Bench-Hard set, and that SGM improves performance over Descriptive Memory commonly adopted in prior works. Code available at: https://github.com/JingbiaoMei/ATM-Bench
Authors: Jan Finke, Wayne Paul Martis, Adrian Schmelter, Lars Erbach, Christian Jestel, Marvin Wiedemann
Abstract: Reliable obstacle avoidance in industrial settings demands 3D scene understanding, but widely used 2D LiDAR sensors perceive only a single horizontal slice of the environment, missing critical obstacles above or below the scan plane. We present a teacher-student framework for vision-based mobile robot navigation that eliminates the need for LiDAR sensors. A teacher policy trained via Proximal Policy Optimization (PPO) in NVIDIA Isaac Lab leverages privileged 2D LiDAR observations that account for the full robot footprint to learn robust navigation. The learned behavior is distilled into a student policy that relies solely on monocular depth maps predicted by a fine-tuned Depth Anything V2 model from four RGB cameras. The complete inference pipeline, comprising monocular depth estimation (MDE), policy execution, and motor control, runs entirely onboard an NVIDIA Jetson Orin AGX mounted on a DJI RoboMaster platform, requiring no external computation for inference. In simulation, the student achieves success rates of 82-96.5%, consistently outperforming the standard 2D LiDAR teacher (50-89%). In real-world experiments, the MDE-based student outperforms the 2D LiDAR teacher when navigating around obstacles with complex 3D geometries, such as overhanging structures and low-profile objects, that fall outside the single scan plane of a 2D LiDAR.
Authors: Jiachun Li, Shaoping Huang, Zhuoran Jin, Chenlong Zhang, Pengfei Cao, Yubo Chen, Kang Liu, Jun Zhao
Abstract: Recent progress in the reasoning capabilities of multimodal large language models (MLLMs) has empowered them to address more complex tasks such as scientific analysis and mathematical reasoning. Despite their promise, MLLMs' reasoning abilities across different scenarios in real life remain largely unexplored and lack standardized benchmarks for evaluation. To address this gap, we introduce MMR-Life, a comprehensive benchmark designed to evaluate the diverse multimodal multi-image reasoning capabilities of MLLMs across real-life scenarios. MMR-Life consists of 2,646 multiple-choice questions based on 19,108 images primarily sourced from real-world contexts, comprehensively covering seven reasoning types: abductive, analogical, causal, deductive, inductive, spatial, and temporal. Unlike existing reasoning benchmarks, MMR-Life does not rely on domain-specific expertise but instead requires models to integrate information across multiple images and apply diverse reasoning abilities. The evaluation of 37 advanced models highlights the substantial challenge posed by MMR-Life. Even top models like GPT-5 achieve only 58% accuracy and display considerable variance in performance across reasoning types. Moreover, we analyze the reasoning paradigms of existing MLLMs, exploring how factors such as thinking length, reasoning method, and reasoning type affect their performance. In summary, MMR-Life establishes a comprehensive foundation for evaluating, analyzing, and improving the next generation of multimodal reasoning systems.
Authors: Fabian Schmidt, Karol Fedurko, Markus Enzweiler, Abhinav Valada
Abstract: While multimodal large language models (MLLMs) provide advanced reasoning for autonomous driving, translating their discrete semantic knowledge into continuous trajectories remains a fundamental challenge. Existing methods often rely on unimodal planning heads that inherently limit their ability to represent multimodal driving behavior. Furthermore, most generative approaches frequently condition on one-hot encoded actions, discarding the nuanced navigational uncertainty critical for complex scenarios. To resolve these limitations, we introduce LAD-Drive, a generative framework that structurally disentangles high-level intention from low-level spatial planning. LAD-Drive employs an action decoder to infer a probabilistic meta-action distribution, establishing an explicit belief state that preserves the nuanced intent typically lost by one-hot encodings. This distribution, fused with the vehicle's kinematic state, conditions an action-aware diffusion decoder that utilizes a truncated denoising process to refine learned motion anchors into safe, kinematically feasible trajectories. Extensive evaluations on the LangAuto benchmark demonstrate that LAD-Drive achieves state-of-the-art results, outperforming competitive baselines by up to 59% in Driving Score while significantly reducing route deviations and collisions. We will publicly release the code and models on https://github.com/iis-esslingen/lad-drive.
Authors: Siting Wang, Xiaofeng Wang, Zheng Zhu, Minnan Pei, Xinyu Cui, Cheng Deng, Jian Zhao, Guan Huang, Haifeng Zhang, Jun Wang
Abstract: Flow-based vision-language-action (VLA) models excel in embodied control but suffer from intractable likelihoods during multi-step sampling, hindering online reinforcement learning. We propose \textbf{\textit{$\boldsymbol{\pi}$-StepNFT}} (Step-wise Negative-aware Fine-Tuning), a critic-and-likelihood-free framework that requires only a single forward pass per optimization step and eliminates auxiliary value networks. We identify that wider exploration spaces necessitate finer-grained, step-wise guidance for alignment. Empirically, $\pi$-StepNFT unlocks latent potential on LIBERO with competitive few-shot robustness. Moreover, it achieves superior generalization on ManiSkill, outperforming value-based baselines in OOD scenarios by preventing overfitting to multimodal features. This property offers a scalable solution promising for complex real-world applications.
Authors: Chuong Huynh, Manh Luong, Abhinav Shrivastava
Abstract: Multimodal retrieval is the task of aggregating information from queries across heterogeneous modalities to retrieve desired targets. State-of-the-art multimodal retrieval models can understand complex queries, yet they are typically limited to two modalities: text and vision. This limitation impedes the development of universal retrieval systems capable of comprehending queries that combine more than two modalities. To advance toward this goal, we present OmniRet, the first retrieval model capable of handling complex, composed queries spanning three key modalities: text, vision, and audio. Our OmniRet model addresses two critical challenges for universal retrieval: computational efficiency and representation fidelity. First, feeding massive token sequences from modality-specific encoders to Large Language Models (LLMs) is computationally inefficient. We therefore introduce an attention-based resampling mechanism to generate compact, fixed-size representations from these sequences. Second, compressing rich omni-modal data into a single embedding vector inevitably causes information loss and discards fine-grained details. We propose Attention Sliced Wasserstein Pooling to preserve these fine-grained details, leading to improved omni-modal representations. OmniRet is trained on an aggregation of approximately 6 million query-target pairs spanning 30 datasets. We benchmark our model on 13 retrieval tasks and a MMEBv2 subset. Our model demonstrates significant improvements on composed query, audio and video retrieval tasks, while achieving on-par performance with state-of-the-art models on others. Furthermore, we curate a new Audio-Centric Multimodal Benchmark (ACM). This new benchmark introduces two critical, previously missing tasks-composed audio retrieval and audio-visual retrieval to more comprehensively evaluate a model's omni-modal embedding capacity.
Authors: Jiahao Huang, Fengyan Lin, Xuechao Yang, Chen Feng, Kexin Zhu, Xu Yang, Zhide Chen
Abstract: The development of affective multimodal language models (MLMs) has long been constrained by a gap between low-level perception and high-level interaction, leading to fragmented affective capabilities and limited generalization. To bridge this gap, we propose a cognitively inspired three-level hierarchy that organizes affective tasks according to their cognitive depth-perception, understanding, and interaction-and provides a unified conceptual foundation for advancing affective modeling. Guided by this hierarchy, we introduce Nano-EmoX, a small-scale multitask MLM, and P2E (Perception-to-Empathy), a curriculum-based training framework. Nano-EmoX integrates a suite of omni-modal encoders, including an enhanced facial encoder and a fusion encoder, to capture key multimodal affective cues and improve cross-task transferability. The outputs are projected into a unified language space via heterogeneous adapters, empowering a lightweight language model to tackle diverse affective tasks. Concurrently, P2E progressively cultivates emotional intelligence by aligning rapid perception with chain-of-thought-driven empathy. To the best of our knowledge, Nano-EmoX is the first compact MLM (2.2B) to unify six core affective tasks across all three hierarchy levels, achieving state-of-the-art or highly competitive performance across multiple benchmarks, demonstrating excellent efficiency and generalization.
Authors: Yiyan Xu, Ruoxuan Xia, Wuqiang Zheng, Fengbin Zhu, Wenjie Wang, Fuli Feng
Abstract: With the rapid growth of online video consumption, video advertising has become increasingly dominant in the digital advertising landscape. Yet diverse users and viewing contexts makes one-size-fits-all ad creatives insufficient for consistent effectiveness, underlining the importance of personalization. In practice, most personalized video advertising systems follow a retrieval-based paradigm, selecting the optimal one from a small set of professionally pre-produced creatives for each user. Such static and finite inventories limits both the granularity and the timeliness of personalization, and prevents the creatives from being continuously refined based on online user feedback. Recent advances in generative AI make it possible to move beyond retrieval toward optimizing video creatives in a continuous space at serving time. In this light, we propose NextAds, a generation-based paradigm for next-generation personalized video advertising, and conceptualize NextAds with four core components. To enable comparable research progress, we formulate two representative tasks: personalized creative generation and personalized creative integration, and introduce corresponding lightweight benchmarks. To assess feasibility, we instantiate end-to-end pipelines for both tasks and conduct initial exploratory experiments, demonstrating that GenAI can generate and integrate personalized creatives with encouraging performance. Moreover, we discuss the key challenges and opportunities under this paradigm, aiming to provide actionable insights for both researchers and practitioners and to catalyze progress in personalized video advertising.
Authors: Han Xue, Nan Min, Xiaotong Liu, Wendi Chen, Yuan Fang, Jun Lv, Cewu Lu, Chuan Wen
Abstract: The adoption of fisheye cameras in robotic manipulation, driven by their exceptionally wide Field of View (FoV), is rapidly outpacing a systematic understanding of their downstream effects on policy learning. This paper presents the first comprehensive empirical study to bridge this gap, rigorously analyzing the properties of wrist-mounted fisheye cameras for imitation learning. Through extensive experiments in both simulation and the real world, we investigate three critical research questions: spatial localization, scene generalization, and hardware generalization. Our investigation reveals that: (1) The wide FoV significantly enhances spatial localization, but this benefit is critically contingent on the visual complexity of the environment. (2) Fisheye-trained policies, while prone to overfitting in simple scenes, unlock superior scene generalization when trained with sufficient environmental diversity. (3) While naive cross-camera transfer leads to failures, we identify the root cause as scale overfitting and demonstrate that hardware generalization performance can be improved with a simple Random Scale Augmentation (RSA) strategy. Collectively, our findings provide concrete, actionable guidance for the large-scale collection and effective use of fisheye datasets in robotic learning. More results and videos are available on https://robo-fisheye.github.io/
Authors: Junke Wang, Zuxuan Wu, Dongdong Chen, Chong Luo, Xiyang Dai, Lu Yuan, Yu-Gang Jiang
Abstract: Visual Object Tracking (VOT) aims to estimate the positions of target objects in a video sequence, which is an important vision task with various real-world applications. Depending on whether the initial states of target objects are specified by provided annotations in the first frame or the categories, VOT could be classified as instance tracking (e.g., SOT and VOS) and category tracking (e.g., MOT, MOTS, and VIS) tasks. Different definitions have led to divergent solutions for these two types of tasks, resulting in redundant training expenses and parameter overhead. In this paper, combing the advantages of the best practices developed in both communities, we propose a novel tracking-with-detection paradigm, where tracking supplements appearance priors for detection and detection provides tracking with candidate bounding boxes for the association. Equipped with such a design, a unified tracking model, OmniTracker, is further presented to resolve all the tracking tasks with a fully shared network architecture, model weights, and inference pipeline, eliminating the need for task-specific architectures and reducing redundancy in model parameters. We conduct extensive experimentation on seven prominent tracking datasets of different tracking tasks, including LaSOT, TrackingNet, DAVIS16-17, MOT17, MOTS20, and YTVIS19, and demonstrate that OmniTracker achieves on-par or even better results than both task-specific and unified tracking models.
Authors: Xinge Yang, Qiang Fu, Yunfeng Nie, Wolfgang Heidrich
Abstract: Classical lens design minimizes optical aberrations to produce sharp images, but is typically decoupled from downstream computer vision tasks. Existing end-to-end optical design learns optical encoding through joint optimization, but often suffers from an unstable training process. We propose task-driven lens design, a new optimization philosophy for joint optics-network systems. We freeze the pretrained vision model and optimize only the lens so that the image formation better fits the model's feature preferences. This network-frozen setting yields a low-dimensional and stable optimization process, enabling lens design from scratch without human intervention, thereby exploring a broader design space. Multiple computer vision experiments show that TaskLenses outperform classical ImagingLenses with the same or even fewer elements. Our analysis reveals that the learned optics exhibit long-tailed point spread functions, better preserving preferred structural cues when aberrations cannot be fully corrected. These results highlight task-driven design as a practical route for optical lenses that are compatible with modern vision models, and also inspire new optical design objectives beyond traditional aberration minimization.
Authors: Zhihang Zhong, Yiming Zhang, Wei Wang, Xiao Sun, Yu Qiao, Gurunandan Krishnan, Sizhuo Ma, Jian Wang
Abstract: Existing video frame interpolation (VFI) methods blindly predict where each object is at a specific timestep t ("time indexing"), which struggles to predict precise object movements. Given two images of a baseball, there are infinitely many possible trajectories: accelerating or decelerating, straight or curved. This often results in blurry frames as the method averages out these possibilities. Instead of forcing the network to learn this complicated time-to-location mapping implicitly, we provide the network with an explicit hint on how far the object has traveled between start and end frames, a novel approach termed "distance indexing". This method offers a clearer learning goal for models, reducing the uncertainty tied to object speeds. Moreover, even with this extra guidance, objects can still be blurry especially when they are equally far from both input frames, due to the directional ambiguity in long-range motion. To solve this, we propose an iterative reference-based estimation strategy that breaks down a long-range prediction into several short-range steps. When integrating our plug-and-play strategies into state-of-the-art learning-based models, they exhibit markedly superior perceptual quality in arbitrary time interpolations, using a uniform distance indexing map in the same format as time indexing without requiring extra computation. Furthermore, we demonstrate that if additional latency is acceptable, a continuous map estimator can be employed to compute a pixel-wise dense distance indexing using multiple nearby frames. Combined with efficient multi-frame refinement, this extension can further disambiguate complex motion, thus enhancing performance both qualitatively and quantitatively. Additionally, the ability to manually specify distance indexing allows for independent temporal manipulation of each object, providing a novel tool for video editing tasks such as re-timing.
Authors: Zhaopeng Gu, Bingke Zhu, Guibo Zhu, Yingying Chen, Hao Li, Ming Tang, Jinqiao Wang
Abstract: Zero-shot anomaly detection (ZSAD) methods entail detecting anomalies directly without access to any known normal or abnormal samples within the target item categories. Existing approaches typically rely on the robust generalization capabilities of multimodal pretrained models, computing similarities between manually crafted textual features representing "normal" or "abnormal" semantics and image features to detect anomalies and localize anomalous patches. However, the generic descriptions of "abnormal" often fail to precisely match diverse types of anomalies across different object categories. Additionally, computing feature similarities for single patches struggles to pinpoint specific locations of anomalies with various sizes and scales. To address these issues, we propose a novel ZSAD method called FiLo, comprising two components: adaptively learned Fine-Grained Description (FG-Des) and position-enhanced High-Quality Localization (HQ-Loc). FG-Des introduces fine-grained anomaly descriptions for each category using Large Language Models (LLMs) and employs adaptively learned textual templates to enhance the accuracy and interpretability of anomaly detection. HQ-Loc, utilizing Grounding DINO for preliminary localization, position-enhanced text prompts, and Multi-scale Multi-shape Cross-modal Interaction (MMCI) module, facilitates more accurate localization of anomalies of different sizes and shapes. Experimental results on datasets like MVTec and VisA demonstrate that FiLo significantly improves the performance of ZSAD in both detection and localization, achieving state-of-the-art performance with an image-level AUC of 83.9% and a pixel-level AUC of 95.9% on the VisA dataset. Code is available at https://github.com/CASIA-IVA-Lab/FiLo.
Authors: Zhuolin He, Xinrun Li, Jiacheng Tang, Shoumeng Qiu, Wenfu Wang, Xiangyang Xue, Jian Pu
Abstract: Conventional camera-based 3D object detectors in autonomous driving are limited to recognizing a predefined set of objects, which poses a safety risk when encountering novel or unseen objects in real-world scenarios. To address this limitation, we present OS-Det3D, a two-stage training framework designed for camera-based open-set 3D object detection. In the first stage, our proposed 3D object discovery network (ODN3D) uses geometric cues from LiDAR point clouds to generate class-agnostic 3D object proposals, each of which are assigned a 3D objectness score. This approach allows the network to discover objects beyond known categories, allowing for the detection of unfamiliar objects. However, due to the absence of class constraints, ODN3D-generated proposals may include noisy data, particularly in cluttered or dynamic scenes. To mitigate this issue, we introduce a joint selection (JS) module in the second stage. The JS module uses both camera bird's eye view (BEV) feature responses and 3D objectness scores to filter out low-quality proposals, yielding high-quality pseudo ground truth for unknown objects. OS-Det3D significantly enhances the ability of camera 3D detectors to discover and identify unknown objects while also improving the performance on known objects, as demonstrated through extensive experiments on the nuScenes and KITTI datasets.
Authors: Ricardo Pizarro, Roberto Valle, Rafael Barea, Jose M. Buenaposada, Luis Baumela, Luis Miguel Bergasa
Abstract: We address the task of identifying distracted driving by analyzing in-car videos using efficient transformers. Although transformer models have achieved outstanding performance in human action recognition tasks, their high computational costs limit their application onboard a vehicle. We introduce POGUISE+, a multi-task video transformer that, given an input clip, predicts the distracted driving action, the driver's pose, and the interacting object. Our enhanced features for token selection are specifically adapted to driver actions by leveraging information about object interaction and the driver's pose. With POGUISE+, we significantly reduce the model's computational demands while maintaining or improving baseline accuracy across various computational budgets. Additionally, to evaluate our model's performance in real-world scenarios, we have developed benchmarks on a Jetson computing platform, demonstrating its effectiveness across different configurations and computational budgets. Our model outperforms current state-of-the-art results on the Drive&Act, 100-Driver, and 3MDAD datasets, while having superior efficiency compared to existing video transformer-based methods.
Authors: Md Touhidul Islam, Imran Kabir, Elena Ariel Pearce, Md Alimoor Reza, Syed Masum Billah
Abstract: This paper introduces a dataset for improving real-time object recognition systems to aid blind and low-vision (BLV) individuals in navigation tasks. The dataset comprises 21 videos of BLV individuals navigating outdoor spaces, and a taxonomy of 90 objects crucial for BLV navigation, refined through a focus group study. We also provide object labeling for the 90 objects across 31 video segments created from the 21 videos. A deeper analysis reveals that most contemporary datasets used in training computer vision models contain only a small subset of the taxonomy in our dataset. Preliminary evaluation of state-of-the-art computer vision models on our dataset highlights shortcomings in accurately detecting key objects relevant to BLV navigation, emphasizing the need for specialized datasets. We make our dataset publicly available, offering valuable resources for developing more inclusive navigation systems for BLV individuals.
Authors: Taorong Liu, Zhen Zhang, Liang Liao, Jing Xiao, Chia-Wen Lin
Abstract: Test-time domain adaption (TTDA) for semantic segmentation aims to adapt a segmentation model trained on a source domain to a target domain for inference on-the-fly, where both efficiency and effectiveness are critical. However, existing TTDA methods either rely on costly frame-wise optimization or assume unrealistic domain shifts, resulting in poor adaptation efficiency and continuous semantic ambiguities. To address these challenges, we propose a real-time framework for TTDA semantic segmentation, called Dynamic Ambiguity-Wise Adaptation (DAWA), which adaptively detects domain shifts and dynamically adjusts the learning strategies to mitigate continuous ambiguities in the test time. Specifically, we introduce the Dynamic Ambiguous Patch Mask (DAP Mask) strategy, which dynamically identifies and masks highly disturbed regions to prevent error accumulation in ambiguous classes. Furthermore, we present the Dynamic Ambiguous Class Mix (DAC Mix) strategy that leverages vision-language models to group semantically similar classes and augment the target domain with a meta-ambiguous class buffer. Extensive experiments on widely used TTDA benchmarks demonstrate that DAWA consistently outperforms state-of-the-art methods, while maintaining real-time inference speeds of approximately 40 FPS.
Authors: Wei Peng, Tian Xia, Fabio De Sousa Ribeiro, Tomas Bosschieter, Ehsan Adeli, Qingyu Zhao, Ben Glocker, Kilian M. Pohl
Abstract: The number of samples in structural brain MRI studies is often too small to properly train deep learning models. Generative models show promise in addressing this issue by effectively learning the data distribution and generating high-fidelity MRI. However, they struggle to produce diverse, high-quality data outside the distribution defined by the training data. One way to address this issue is to use causal models developed for 3D volume counterfactuals. However, accurately modeling causality in high-dimensional spaces is challenging, so these models generally generate 3D brain MRIs of lower quality. To address these challenges, we propose a two-stage method that constructs a Structural Causal Model (SCM) within the latent space. In the first stage, we employ a VQ-VAE to learn a compact embedding of the MRI volume. Subsequently, we integrate our causal model into this latent space and execute a three-step counterfactual procedure using a closed-form Generalized Linear Model (GLM). Our experiments conducted on real-world high-resolution MRI data (1 mm) provided by the Alzheimer's Disease Neuroimaging Initiative (ADNI) and the National Consortium on Alcohol and Neurodevelopment in Adolescence (NCANDA) demonstrate that our method can generate high-quality 3D MRI counterfactuals.
Authors: Cheng Lei, Jie Fan, Xinran Li, Tianzhu Xiang, Ao Li, Ce Zhu, Le Zhang
Abstract: Camouflaged Object Segmentation (COS) faces significant challenges due to the scarcity of annotated data, where meticulous pixel-level annotation is both labor-intensive and costly, primarily due to the intricate object-background boundaries. Addressing the core question, "Can COS be effectively achieved in a zero-shot manner without manual annotations for any camouflaged object?" we affirmatively respond and introduce a robust zero-shot COS framework. This framework leverages the inherent local pattern bias of COS and employs a broad semantic feature space derived from salient object segmentation (SOS) for efficient zero-shot transfer. We incorporate an Masked Image Modeling (MIM) based image encoder optimized for Parameter-Efficient Fine-Tuning (PEFT), a Multimodal Large Language Model (M-LLM), and a Multi-scale Fine-grained Alignment (MFA) mechanism. The MIM pre-trained image encoder focuses on capturing essential low-level features, while the M-LLM generates caption embeddings processed alongside these visual cues. These embeddings are precisely aligned using MFA, enabling our framework to accurately interpret and navigate complex semantic contexts. To optimize operational efficiency, we introduce a learnable codebook that represents the M-LLM during inference, significantly reducing computational overhead. Our framework demonstrates its versatility and efficacy through rigorous experimentation, achieving state-of-the-art performance in zero-shot COS with $F_{\beta}^w$ scores of 72.9\% on CAMO and 71.7\% on COD10K. By removing the M-LLM during inference, we achieve an inference speed comparable to that of traditional end-to-end models, reaching 18.1 FPS. Code: https://github.com/AVC2-UESTC/ZSCOS-CaMF
Authors: Lianjun Liu
Abstract: Underwater instance segmentation is a fundamental and critical step in various underwater vision tasks. However, the decline in image quality caused by complex underwater environments presents significant challenges to existing segmentation models. While the state-of-the-art USIS-SAM model has demonstrated impressive performance, it struggles to effectively adapt to feature variations across different channels in addressing issues such as light attenuation, color distortion, and complex backgrounds. This limitation hampers its segmentation performance in challenging underwater scenarios. To address these issues, we propose the MarineVision Adapter (MV-Adapter). This module introduces an adaptive channel attention mechanism that enables the model to dynamically adjust the feature weights of each channel based on the characteristics of underwater images. By adaptively weighting features, the model can effectively handle challenges such as light attenuation, color shifts, and complex backgrounds. Experimental results show that integrating the MV-Adapter module into the USIS-SAM network architecture further improves the model's overall performance, especially in high-precision segmentation tasks. On the USIS10K dataset, the module achieves improvements in key metrics such as mAP, AP50, and AP75 compared to competitive baseline models.
Authors: Jinge Ma, Xiaoyan Zhang, Gautham Vinod, Siddeshwar Raghavan, Jiangpeng He, Fengqing Zhu
Abstract: Food portion estimation is crucial for monitoring health and tracking dietary intake. Image-based dietary assessment, which involves analyzing eating occasion images using computer vision techniques, is increasingly replacing traditional methods such as 24-hour recalls. However, accurately estimating the nutritional content from images remains challenging due to the loss of 3D information when projecting to the 2D image plane. Existing portion estimation methods are challenging to deploy in real-world scenarios due to their reliance on specific requirements, such as physical reference objects, high-quality depth information, or multi-view images and videos. In this paper, we introduce MFP3D, a new framework for accurate food portion estimation using only a single monocular image. Specifically, MFP3D consists of three key modules: (1) a 3D Reconstruction Module that generates a 3D point cloud representation of the food from the 2D image, (2) a Feature Extraction Module that extracts and concatenates features from both the 3D point cloud and the 2D RGB image, and (3) a Portion Regression Module that employs a deep regression model to estimate the food's volume and energy content based on the extracted features. Our MFP3D is evaluated on MetaFood3D dataset, demonstrating its significant improvement in accurate portion estimation over existing methods.
Authors: Zheng Chen, Xun Zhang, Wenbo Li, Renjing Pei, Fenglong Song, Xiongkuo Min, Xiaohong Liu, Xin Yuan, Yong Guo, Yulun Zhang
Abstract: The development of multimodal large language models (MLLMs) enables the evaluation of image quality through natural language descriptions. This advancement allows for more detailed assessments. However, these MLLM-based IQA methods primarily rely on general contextual descriptions, sometimes limiting fine-grained quality assessment. To address this limitation, we introduce a new image quality assessment (IQA) task paradigm, **grounding-IQA**. This paradigm integrates multimodal referring and grounding with IQA to realize more fine-grained quality perception, thereby extending existing IQA. Specifically, grounding-IQA comprises two subtasks: grounding-IQA-description (GIQA-DES) and visual question answering (GIQA-VQA). GIQA-DES involves detailed descriptions with precise locations (e.g., bounding boxes), while GIQA-VQA focuses on quality QA for local regions. To realize grounding-IQA, we construct a corresponding dataset, GIQA-160K, through our proposed automated annotation pipeline. Furthermore, we develop a well-designed benchmark, GIQA-Bench. The benchmark evaluates the grounding-IQA performance from three perspectives: description quality, VQA accuracy, and grounding precision. Experiments demonstrate that our proposed method facilitates the more fine-grained IQA application. Code: https://github.com/zhengchen1999/Grounding-IQA.
Authors: Li-Jun Zhao, Si-Yuan Zhang, Zhen-Duo Chen, Xin Luo, Xin-Shun Xu
Abstract: Current fine-grained classification research primarily focuses on fine-grained feature learning. However, in real-world scenarios, fine-grained data annotation is challenging, and the features and semantics are highly diverse and frequently changing. These issues create inherent barriers between traditional experimental settings and real-world applications, limiting the effectiveness of conventional fine-grained classification methods. Although some recent studies have provided potential solutions to these issues, most of them still rely on limited supervised information and thus fail to offer effective solutions. In this paper, based on theoretical analysis, we propose a novel learning paradigm to break the barriers in fine-grained classification. This paradigm enables the model to progressively learn during inference, thereby leveraging cost-free data at inference time to more accurately represent fine-grained categories and adapt to dynamic semantic changes. On this basis, an efficient EXPloring and EXPloiting strategy and method (EXP2) is designed. Thereinto, useful inference data samples are explored according to class representations and exploited to optimize classifiers. Experimental results demonstrate the general effectiveness of our method, providing guidance for future in-depth understanding and exploration of real-world fine-grained classification.
Authors: Zhaopeng Gu, Bingke Zhu, Guibo Zhu, Yingying Chen, Ming Tang, Jinqiao Wang
Abstract: Anomaly detection methods typically require extensive normal samples from the target class for training, limiting their applicability in scenarios that require rapid adaptation, such as cold start. Zero-shot and few-shot anomaly detection do not require labeled samples from the target class in advance, making them a promising research direction. Existing zero-shot and few-shot approaches often leverage powerful multimodal models to detect and localize anomalies by comparing image-text similarity. However, their handcrafted generic descriptions fail to capture the diverse range of anomalies that may emerge in different objects, and simple patch-level image-text matching often struggles to localize anomalous regions of varying shapes and sizes. To address these issues, this paper proposes the FiLo++ method, which consists of two key components. The first component, Fused Fine-Grained Descriptions (FusDes), utilizes large language models to generate anomaly descriptions for each object category, combines both fixed and learnable prompt templates and applies a runtime prompt filtering method, producing more accurate and task-specific textual descriptions. The second component, Deformable Localization (DefLoc), integrates the vision foundation model Grounding DINO with position-enhanced text descriptions and a Multi-scale Deformable Cross-modal Interaction (MDCI) module, enabling accurate localization of anomalies with various shapes and sizes. In addition, we design a position-enhanced patch matching approach to improve few-shot anomaly detection performance. Experiments on multiple datasets demonstrate that FiLo++ achieves significant performance improvements compared with existing methods. Code will be available at https://github.com/CASIA-IVA-Lab/FiLo.
Authors: Darina Koishigarina, Arnas Uselis, Seong Joon Oh
Abstract: CLIP (Contrastive Language-Image Pretraining) has become a popular choice for various downstream tasks. However, recent studies have questioned its ability to represent compositional concepts effectively. These works suggest that CLIP often acts like a bag-of-words (BoW) model, interpreting images and text as sets of individual concepts without grasping the structural relationships. In particular, CLIP struggles to correctly bind attributes to their corresponding objects when multiple objects are present in an image or text. In this work, we investigate why CLIP exhibits this BoW-like behavior. Our key finding is that CLIP does not lack binding information. Through linear probing, robustness tests with increasing object counts, and conjunctive search experiments, we show that attribute-object bindings are already encoded within CLIP's text and image embeddings. The weakness lies in the cross-modal alignment, which fails to preserve this information. We show it can be accessed cross-modally with a simple linear transformation to text embeddings. This improves CLIP's attribute-object binding performance and confirms that the information was already encoded unimodally. In practice, this means CLIP-based systems can be enhanced with a lightweight linear layer trained on existing embeddings, avoiding costly encoder retraining. The code is available at https://github.com/kdariina/CLIP-not-BoW-unimodally.
Authors: Jack Hong, Shilin Yan, Jiayin Cai, Xiaolong Jiang, Yao Hu, Weidi Xie
Abstract: We introduce WorldSense, the first benchmark to assess the multi-modal video understanding, that simultaneously encompasses visual, audio, and text inputs. In contrast to existing benchmarks, our WorldSense has several features: (i)collaboration of omni-modality, we design the evaluation tasks to feature a strong coupling of audio and video, requiring models to effectively utilize the synergistic perception of omni-modality; (ii)diversity of videos and tasks, WorldSense encompasses a diverse collection of 1,662 audio-visual synchronised videos, systematically categorized into 8 primary domains and 67 fine-grained subcategories to cover the broad scenarios, and 3,172 multi-choice QA pairs across 26 distinct tasks to enable the comprehensive evaluation; (iii)high-quality annotations, all the QA pairs are manually labeled by 80 expert annotators with multiple rounds of correction to ensure quality. Based on our WorldSense, we extensively evaluate various state-of-the-art models. The experimental results indicate that existing models face significant challenges in understanding real-world scenarios (65.1% best accuracy). By analyzing the limitations of current models, we aim to provide valuable insight to guide development of real-world understanding. We hope our WorldSense can provide a platform for evaluating the ability in constructing and understanding coherent contexts from omni-modality.
Authors: {\L}ukasz Staniszewski, Bartosz Cywi\'nski, Franziska Boenisch, Kamil Deja, Adam Dziedzic
Abstract: Novel diffusion models can synthesize photo-realistic images with integrated high-quality text. Surprisingly, we demonstrate through attention activation patching that only less than $1$% of diffusion models' parameters, all contained in attention layers, influence the generation of textual content within the images. Building on this observation, we improve textual generation efficiency and performance by targeting cross and joint attention layers of diffusion models. We introduce several applications that benefit from localizing the layers responsible for textual content generation. We first show that a LoRA-based fine-tuning solely of the localized layers enhances, even more, the general text-generation capabilities of large diffusion models while preserving the quality and diversity of the diffusion models' generations. Then, we demonstrate how we can use the localized layers to edit textual content in generated images. Finally, we extend this idea to the practical use case of preventing the generation of toxic text in a cost-free manner. In contrast to prior work, our localization approach is broadly applicable across various diffusion model architectures, including U-Net (e.g., SDXL and DeepFloyd IF) and transformer-based (e.g., Stable Diffusion 3), utilizing diverse text encoders (e.g., from CLIP to the large language models like T5). Project page available at https://t2i-text-loc.github.io/.
Authors: Anthony Fuller, Yousef Yassin, Daniel G. Kyrollos, Evan Shelhamer, James R. Green
Abstract: ViTs are general and accurate, and address many tasks, but ViTs are slow, and are not always practical when efficiency is key. Existing methods for faster ViTs design hybrid non-ViT architectures, losing generality, or shrink their tokens, sacrificing accuracy. Many non-ViT architectures are both fast and accurate. Yet, without significant modifications, they cannot do what ViTs can: process other input shapes, pre-train by SOTA self-supervised learning, reduce computation by dropping tokens, and more. We make ViTs faster by reducing patch token width while increasing global token width by adding a new Jumbo token. Our wider Jumbo token is processed by its own wider FFN to increase model capacity. Yet our Jumbo FFN is efficient: it processes a single token, for speed, and its parameters are shared across all layers, for memory. Crucially, our Jumbo is attention-only and non-hierarchical, like a plain ViT, so it is simple, scalable, flexible, and compatible with ViT methods new and old. Jumbo improves over ViT baselines with Registers from Nano to Large scales while maintaining speed/throughput on ImageNet-1K (0.1-13%). Jumbo also improves segmentation (1.9-3.1% on ADE20K), MAE pre-training (4.9% linear probing on ImageNet-1K), test-time adaptation (5.2% on ImageNet-C), and time series modeling. Our Jumbo models even achieve better speed-accuracy trade-offs than specialized non-ViT compute-efficient models, while maintaining plain-ViT compatibility for practicality. Code and weights are available: https://github.com/antofuller/jumbo
Authors: Zhibin Lan, Liqiang Niu, Fandong Meng, Jie Zhou, Jinsong Su
Abstract: Universal multimodal embedding models play a critical role in tasks such as interleaved image-text retrieval, multimodal RAG, and multimodal clustering. However, our empirical results indicate that existing LMM-based embedding models trained with the standard InfoNCE loss exhibit a high degree of overlap in similarity distribution between positive and negative pairs, making it challenging to distinguish hard negative pairs effectively. To deal with this issue, we propose a simple yet effective framework that dynamically improves the embedding model's representation learning for negative pairs based on their discriminative difficulty. Within this framework, we train a series of models, named LLaVE, and evaluate them on the MMEB benchmark, which covers 4 meta-tasks and 36 datasets. Experimental results show that LLaVE establishes stronger baselines that achieve state-of-the-art (SOTA) performance while demonstrating strong scalability and efficiency. Specifically, LLaVE-2B surpasses the previous SOTA 7B models, while LLaVE-7B achieves a further performance improvement of 6.2 points. Although LLaVE is trained on image-text data, it can generalize to text-video retrieval tasks in a zero-shot manner and achieve strong performance, demonstrating its remarkable potential for transfer to other embedding tasks.
Authors: Wenxuan Huang, Bohan Jia, Zijie Zhai, Shaosheng Cao, Zheyu Ye, Fei Zhao, Zhe Xu, Xu Tang, Yao Hu, Shaohui Lin
Abstract: DeepSeek-R1-Zero has successfully demonstrated the emergence of reasoning capabilities in LLMs purely through Reinforcement Learning (RL). Inspired by this breakthrough, we explore how RL can be utilized to enhance the reasoning capability of MLLMs. However, direct training with RL struggles to activate complex reasoning capabilities such as questioning and reflection in MLLMs, due to the absence of substantial high-quality multimodal reasoning data. To address this issue, we propose the reasoning MLLM, Vision-R1, to improve multimodal reasoning capability. Specifically, we first construct a high-quality multimodal CoT dataset without human annotations by leveraging an existing MLLM and DeepSeek-R1 through modality bridging and data filtering to obtain a 200K multimodal CoT dataset, Vision-R1-cold dataset. It serves as cold-start initialization data for Vision-R1. To mitigate the optimization challenges caused by overthinking after cold start, we propose Progressive Thinking Suppression Training (PTST) strategy and employ Group Relative Policy Optimization (GRPO) with the hard formatting result reward function to gradually refine the model's ability to learn correct and complex reasoning processes on a 10K multimodal math dataset. Comprehensive experiments show our model achieves an average improvement of $\sim$6% across various multimodal math reasoning benchmarks. Vision-R1-7B achieves a 73.5% accuracy on the widely used MathVista benchmark, which is only 0.4% lower than the leading reasoning model, OpenAI O1. Scaling up the amount of multimodal math data in the RL training, Vision-R1-32B and Vison-R1-72B achieves 76.4% and 78.2% MathVista benchmark scores, respectively. The datasets and code will be released in: https://github.com/Osilly/Vision-R1 .
Authors: Zisheng Chen, Chunwei Wang, Runhui Huang, Hongbin Xu, Xiuwei Chen, Jun Zhou, Jianhua Han, Hang Xu, Xiaodan Liang
Abstract: In this paper, we introduce SemHiTok, a unified image Tokenizer via Semantic-Guided Hierarchical codebook that provides consistent discrete representations for multimodal understanding and generation. Recently, unified image tokenizers have sparked exploration within the research community, which is designed to capture high-level semantic features for understanding and retaining low-level pixel features for generation. Previous works attempt to train a unified image tokenizer by combining loss for semantic distillation and pixel reconstruction. However, due to the differing levels of features prioritized by multimodal understanding and generation, joint training methods face significant challenges in achieving a good trade-off. SemHiTok addresses this challenge through a novel semantic-guided hierarchical codebook, which builds pixel sub-codebooks on a pretrained semantic codebook. This design decouples the semantic and pixel in terms of structure and training strategy, enabling the tokenizer to capture pixel features while retaining its ability to comprehend high-level semantic information. Our experiments demonstrate that SemHiTok achieves leading performance in image reconstruction and multimodal understanding under the LLaVA-v1.5 setting. Further, we develop a unified MLLM with SemHiTok, which exhibits superior performance across multimodal understanding and generation tasks. Extensive experiments confirm our analysis, showing that our unified image tokenizer architecture achieves a better trade-off.
Authors: Zebin You, Jingyang Ou, Xiaolu Zhang, Jun Hu, Jun Zhou, Chongxuan Li
Abstract: Although masked image generation models and masked diffusion models are designed with different motivations and objectives, we observe that they can be unified within a single framework. Building upon this insight, we carefully explore the design space of training and sampling, identifying key factors that contribute to both performance and efficiency. Based on the improvements observed during this exploration, we develop our model, referred to as \textbf{eMIGM}. Empirically, eMIGM demonstrates strong performance on ImageNet generation, as measured by Fr\'echet Inception Distance (FID). In particular, on ImageNet $256\times256$, with similar number of function evaluations (NFEs) and model parameters, eMIGM outperforms the seminal VAR. Moreover, as NFE and model parameters increase, eMIGM achieves performance comparable to the state-of-the-art continuous diffusion model REPA while requiring less than 45\% of the NFE. Additionally, on ImageNet $512\times512$, eMIGM outperforms the strong continuous diffusion model EDM2. Code is available at https://github.com/ML-GSAI/eMIGM.
Authors: Ouxiang Li, Yuan Wang, Xinting Hu, Houcheng Jiang, Yanbin Hao, Fuli Feng
Abstract: Erasing concepts from large-scale text-to-image (T2I) diffusion models has become increasingly crucial due to the growing concerns over copyright infringement, offensive content, and privacy violations. In scalable applications, fine-tuning-based methods are time-consuming to precisely erase multiple target concepts, while real-time editing-based methods often degrade the generation quality of non-target concepts due to conflicting optimization objectives. To address this dilemma, we introduce SPEED, an efficient concept erasure approach that directly edits model parameters. SPEED searches for a null space, a model editing space where parameter updates do not affect non-target concepts, to achieve scalable and precise erasure. To facilitate accurate null space optimization, we incorporate three complementary strategies: Influence-based Prior Filtering (IPF) to selectively retain the most affected non-target concepts, Directed Prior Augmentation (DPA) to enrich the filtered retain set with semantically consistent variations, and Invariant Equality Constraints (IEC) to preserve key invariants during the T2I generation process. Extensive evaluations across multiple concept erasure tasks demonstrate that SPEED consistently outperforms existing methods in non-target preservation while achieving efficient and high-fidelity concept erasure, successfully erasing 100 concepts within only 5 seconds. Our code and models are available at: https://github.com/Ouxiang-Li/SPEED.
Authors: Fei Wang, Tingting Zhang, Wei Xi, Han Ding, Ge Wang, Di Zhang, Yuanhao Cui, Fan Liu, Jinsong Han, Jie Xu, Tony Xiao Han
Abstract: Wi-Fi sensing has emerged as a powerful non-intrusive technology for recognizing human activities, monitoring vital signs, and enabling context-aware applications using commercial wireless devices. However, the performance of Wi-Fi sensing often degrades when applied to new users, devices, or environments due to significant domain shifts. To address this challenge, researchers have proposed a wide range of generalization techniques aimed at enhancing the robustness and adaptability of Wi-Fi sensing systems. In this survey, we provide a comprehensive and structured review of over 200 papers published since 2015, categorizing them according to the Wi-Fi sensing pipeline: experimental setup, signal preprocessing, feature learning, and model deployment. We analyze key techniques, including signal preprocessing, domain adaptation, meta-learning, metric learning, data augmentation, cross-modal alignment, federated learning, and continual learning. Furthermore, we summarize publicly available datasets across various tasks, such as activity recognition, user identification, indoor localization, and pose estimation, and provide insights into their domain diversity. We also discuss emerging trends and future directions, including large-scale pretraining, integration with multimodal foundation models, and continual deployment. To foster community collaboration, we introduce the Sensing Dataset Platform (SDP) (http://www.sdp8.org/) for sharing datasets and models. This survey aims to serve as a valuable reference and practical guide for researchers and practitioners dedicated to improving the generalizability of Wi-Fi sensing systems.
URLs: http://www.sdp8.org/)
Authors: Taeksoo Kim, Hanbyul Joo
Abstract: We present a target-aware video diffusion model that generates videos from an input image, in which an actor interacts with a specified target while performing a desired action. The target is defined by a segmentation mask, and the action is described through a text prompt. Our key motivation is to incorporate target awareness into video generation, enabling actors to perform directed actions on designated objects. This enables video diffusion models to act as motion planners, producing plausible predictions of human-object interactions by leveraging the priors of large-scale video generative models. We build our target-aware model by extending a baseline model to incorporate the target mask as an additional input. To enforce target awareness, we introduce a special token that encodes the target's spatial information within the text prompt. We then fine-tune the model with our curated dataset using an additional cross-attention loss that aligns the cross-attention maps associated with this token with the input target mask. To further improve performance, we selectively apply this loss to the most semantically relevant attention regions and transformer blocks. Experimental results show that our target-aware model outperforms existing solutions in generating videos where actors interact accurately with the specified targets. We further demonstrate its efficacy in two downstream applications: zero-shot 3D HOI motion synthesis with physical plausibility and long-term video content creation.
Authors: Linfeng Tang, Yeda Wang, Meiqi Gong, Zizhuo Li, Yuxin Deng, Xunpeng Yi, Chunyu Li, Han Xu, Hao Zhang, Jiayi Ma
Abstract: Compared to images, videos better reflect real-world acquisition and possess valuable temporal cues. However, existing multi-sensor fusion research predominantly integrates complementary context from multiple images rather than videos due to the scarcity of large-scale multi-sensor video datasets, limiting research in video fusion and the inherent difficulty of jointly modeling spatial and temporal dependencies in a unified framework. To this end, we construct M3SVD, a benchmark dataset with $220$ temporally synchronized and spatially registered infrared-visible videos comprising $153,797$ frames, bridging the data gap. Secondly, we propose VideoFusion, a multi-modal video fusion model that exploits cross-modal complementarity and temporal dynamics to generate spatio-temporally coherent videos from multi-modal inputs. Specifically, 1) a differential reinforcement module is developed for cross-modal information interaction and enhancement, 2) a complete modality-guided fusion strategy is employed to adaptively integrate multi-modal features, and 3) a bi-temporal co-attention mechanism is devised to dynamically aggregate forward-backward temporal contexts to reinforce cross-frame feature representations. Experiments reveal that VideoFusion outperforms existing image-oriented fusion paradigms in sequences, effectively mitigating temporal inconsistency and interference. Project and M3SVD: https://github.com/Linfeng-Tang/VideoFusion.
Authors: Shunsuke Sakai, Xiangteng He, Chunzhi Gu, Leonid Sigal, Tatsuhito Hasegawa
Abstract: Despite the remarkable success, recent reconstruction-based anomaly detection (AD) methods via diffusion modeling still involve fine-grained noise-strength tuning and computationally expensive multi-step denoising, leading to a fundamental tension between fidelity and efficiency. In this paper, we propose InvAD, a novel inversion-based anomaly detection approach ("detection via noising in latent space") that circumvents explicit reconstruction. Importantly, we contend that the limitations in prior reconstruction-based methods originate from the prevailing "detection via denoising in RGB space" paradigm. To address this, we model AD under a reconstruction-free formulation, which directly infers the final latent variable corresponding to the input image via DDIM inversion, and then measures the deviation based on the known prior distribution for anomaly scoring. Specifically, in approximating the original probability flow ODE using the Euler method, we enforce only a few inversion steps to noise the clean image to pursue inference efficiency. As the added noise is adaptively derived with the learned diffusion model, the original features for the clean testing image can still be leveraged to yield high detection accuracy. We perform extensive experiments and detailed analyses across four widely used industrial and medical AD benchmarks under the unsupervised unified setting to demonstrate the effectiveness of our model, achieving state-of-the-art AD performance and approximately 2x inference-time speedup without diffusion distillation.
Authors: Xiaochen Wei, Weiwei Guo, Wenxian Yu, Feiming Wei, Dongying Li
Abstract: Multimodal remote sensing image registration aligns images from different sensors for data fusion and analysis. However, existing methods often struggle to extract modality-invariant features when faced with large nonlinear radiometric differences, such as those between SAR and optical images. To address these challenges, we propose OSDM-MReg, a novel multimodal image registration framework that bridges the modality gap through image-to-image translation. Specifically, we introduce a one-step unaligned target-guided conditional diffusion model (UTGOS-CDM) to translate source and target images into a unified representation domain. Unlike traditional conditional DDPM that require hundreds of iterative steps for inference, our model incorporates a novel inverse translation objective during training to enable direct prediction of the translated image in a single step at test time, significantly accelerating the registration process. After translation, we design a multimodal multiscale registration network (MM-Reg) that extracts and fuses both unimodal and translated multimodal images using the proposed multimodal fusion strategy, enhancing the robustness and precision of alignment across scales and modalities. Extensive experiments on the OSdataset demonstrate that OSDM-MReg achieves superior registration accuracy compared to state-of-the-art methods.
Authors: Shamim Rahim Refat, Ziyan Shirin Raha, Shuvashis Sarker, Faika Fairuj Preotee, MD. Musfikur Rahman, Tashreef Muhammad, Mohammad Shafiul Alam
Abstract: Diabetic retinopathy is a severe eye condition caused by diabetes where the retinal blood vessels get damaged and can lead to vision loss and blindness if not treated. Early and accurate detection is key to intervention and stopping the disease progressing. For addressing this disease properly, this paper presents a comprehensive approach for automated diabetic retinopathy detection by proposing a new hybrid deep learning model called VR-FuseNet. Diabetic retinopathy is a major eye disease and leading cause of blindness especially among diabetic patients so accurate and efficient automated detection methods are required. To address the limitations of existing methods including dataset imbalance, diversity and generalization issues this paper presents a hybrid dataset created from five publicly available diabetic retinopathy datasets. Essential preprocessing techniques such as SMOTE for class balancing and CLAHE for image enhancement are applied systematically to the dataset to improve the robustness and generalizability of the dataset. The proposed VR-FuseNet model combines the strengths of two state-of-the-art convolutional neural networks, VGG19 which captures fine-grained spatial features and ResNet50V2 which is known for its deep hierarchical feature extraction. This fusion improves the diagnostic performance and achieves an accuracy of 91.824%. The model outperforms individual architectures on all performance metrics demonstrating the effectiveness of hybrid feature extraction in Diabetic Retinopathy classification tasks. To make the proposed model more clinically useful and interpretable this paper incorporates multiple XAI techniques. These techniques generate visual explanations that clearly indicate the retinal features affecting the model's prediction such as microaneurysms, hemorrhages and exudates so that clinicians can interpret and validate.
Authors: Alex Hoi Hang Chan, Otto Brookes, Urs Waldmann, Hemal Naik, Iain D. Couzin, Majid Mirmehdi, No\"el Adiko Houa, Emmanuelle Normand, Christophe Boesch, Lukas Boesch, Mimi Arandjelovic, Hjalmar K\"uhl, Tilo Burghardt, Fumihiro Kano
Abstract: Computer vision methods have demonstrated considerable potential to streamline ecological and biological workflows, with a growing number of datasets and models becoming available to the research community. However, these resources focus predominantly on evaluation using machine learning metrics, with relatively little emphasis on how their application impacts downstream analysis. We argue that models should be evaluated using application-specific metrics that directly represent model performance in the context of its final use case. To support this argument, we present two disparate case studies: (1) estimating chimpanzee abundance and density with camera trap distance sampling when using a video-based behaviour classifier and (2) estimating head rotation in pigeons using a 3D posture estimator. We show that even models with strong machine learning performance (e.g., 87% mAP) can yield data that leads to discrepancies in abundance estimates compared to expert-derived data. Similarly, the highest-performing models for posture estimation do not produce the most accurate inferences of gaze direction in pigeons. Motivated by these findings, we call for researchers to integrate application-specific metrics in ecological/biological datasets, allowing for models to be benchmarked in the context of their downstream application and to facilitate better integration of models into application workflows.
Authors: Ning Cheng, Jinan Xu, Jialing Chen, Bin Fang, Wenjuan Han
Abstract: This paper explores the challenges of integrating tactile sensing into intelligent systems for multimodal reasoning, particularly in enabling commonsense reasoning about the open-ended physical world. We identify two key challenges: modality discrepancy, where existing large touch-language models often treat touch as a mere sub-modality of language, and open-ended tactile data scarcity, where current datasets lack the diversity, open-endness and complexity needed for reasoning. To overcome these challenges, we introduce SToLa, a Self-Adaptive Touch-Language framework. SToLa utilizes Mixture of Experts (MoE) to dynamically process, unify, and manage tactile and language modalities, capturing their unique characteristics. Crucially, we also present a comprehensive tactile commonsense reasoning dataset and benchmark featuring free-form questions and responses, 8 physical properties, 4 interactive characteristics, and diverse commonsense knowledge. Experiments show SToLa exhibits competitive performance compared to existing models on the PhysiCLeAR benchmark and self-constructed datasets, proving the effectiveness of the Mixture of Experts architecture in multimodal management and the performance advantages for open-scenario tactile commonsense reasoning tasks.
Authors: Zequn Xie, Haoming Ji, Chengxuan Li, Lingwei Meng
Abstract: Text-to-image person search aims to identify an individual based on a text description. To reduce data collection costs, large-scale text-image datasets are created from co-occurrence pairs found online. However, this can introduce noise, particularly mismatched pairs, which degrade retrieval performance. Existing methods often focus on negative samples, which amplify this noise. To address these issues, we propose the Dynamic Uncertainty and Relational Alignment (DURA) framework, which includes the Key Feature Selector (KFS) and a new loss function, Dynamic Softmax Hinge Loss (DSH-Loss). KFS captures and models noise uncertainty, improving retrieval reliability. The bidirectional evidence from cross-modal similarity is modeled as a Dirichlet distribution, enhancing adaptability to noisy data. DSH adjusts the difficulty of negative samples to improve robustness in noisy environments. Our experiments on three datasets show that the method offers strong noise resistance and improves retrieval performance in both low- and high-noise scenarios.
Authors: Jie Li, Shengwei Tian, Long Yu, Xin Ning
Abstract: The Chamfer Distance (CD) is a cornerstone objective function for point cloud completion, yet its inherent symmetric weighting mechanism limits the quality of the generated results. By penalizing local detail deviations and global coverage deficiencies equally, standard CD often causes structural defects such as point aggregation and incomplete spatial structures. We introduce the Flexible-weighted Chamfer Distance (FCD), which decouples CD into local precision and global completeness sub-objectives. FCD employs an asymmetric weighting strategy that prioritizes global structural integrity, steering the optimization away from sub-optimal solutions. As a plug-and-play module with negligible overhead, extensive experiments on state-of-the-art networks demonstrate that FCD significantly enhances global distribution metrics while preserving local precision. Specifically, on the ShapeNet55 benchmark using AdaPoinTr, FCD reduces the Density-aware Chamfer Distance (DCD) by approximately 12.4% (from 0.613 to 0.537), effectively mitigating point clustering. Similarly, on the PCN dataset, the proposed method reduces the Earth Mover's Distance (EMD) from 23.79 to 21.40, demonstrating superior global uniformity compared to the standard CD baseline. Furthermore, FCD demonstrates excellent generalization. When applied to diverse tasks and datasets, including real-world scans (KITTI), industrial components (ABC), and point cloud upsampling (PU-GAN), it yields significant quantitative gains and produces visually more uniform and structurally complete point clouds. These results underscore FCD's potential as a versatile objective function for the broader point cloud generation domain.
Authors: Ziwei Zheng, Michael Yang, Jack Hong, Chenxiao Zhao, Guohai Xu, Le Yang, Chao Shen, Xing Yu
Abstract: Large Vision-Language Models excel at multimodal understanding but struggle to deeply integrate visual information into their predominantly text-based reasoning processes, a key challenge in mirroring human cognition. To address this, we introduce DeepEyes, a model that learns to "think with images", trained end-to-end with reinforcement learning without requiring pre-collected reasoning data for cold-start supervised fine-tuning (SFT). Notably, this ability emerges natively, leveraging the model's own grounding capability as an intrinsic function rather than relying on external specialized models or APIs. We enable this capability through active perception, where the model learns to strategically ground its reasoning in visual information, guided by a tailored data selection and reward strategy. DeepEyes achieves significant performance gains on general perception and reasoning benchmarks and also demonstrates improvement in grounding, hallucination, and mathematical reasoning tasks. Interestingly, we observe the distinct evolution of active perception from initial exploration to efficient and accurate exploitation, and diverse thinking patterns that closely mirror human visual reasoning processes. Code is available at https://github.com/Visual-Agent/DeepEyes.
Authors: Conghao Xiong, Zhengrui Guo, Zhe Xu, Yifei Zhang, Raymond Kai-Yu Tong, Si Yong Yeo, Hao Chen, Joseph J. Y. Sung, Irwin King
Abstract: Few-shot Whole Slide Image (WSI) classification is severely hampered by overfitting. We argue that this is not merely a data-scarcity issue but a fundamentally geometric problem. Grounded in the manifold hypothesis, our analysis shows that features from pathology foundation models exhibit a low-dimensional manifold geometry that is easily perturbed by downstream models. This insight reveals a key potential issue in downstream multiple instance learning models: linear layers are geometry-agnostic and, as we show empirically, can distort the manifold geometry of the features. To address this, we propose the Manifold Residual (MR) block, a plug-and-play module that is explicitly geometry-aware. The MR block reframes the linear layer as residual learning and decouples it into two pathways: (1) a fixed, random matrix serving as a geometric anchor that approximately preserves topology while also acting as a spectral shaper to sharpen the feature spectrum; and (2) a trainable, low-rank residual pathway that acts as a residual learner for task-specific adaptation, with its structural bottleneck explicitly mirroring the low effective rank of the features. This decoupling imposes a structured inductive bias and reduces learning to a simpler residual fitting task. Through extensive experiments, we demonstrate that our approach achieves state-of-the-art results with significantly fewer parameters, offering a new paradigm for few-shot WSI classification. Code is available in https://github.com/BearCleverProud/MR-Block.
Authors: Tanqiu Jiang, Jiacheng Liang, Rongyi Zhu, Jiawei Zhou, Fenglong Ma, Ting Wang
Abstract: Large vision-language models (VLMs) are highly vulnerable to multimodal jailbreak attacks that exploit visual-textual interactions to bypass safety guardrails. In this paper, we present DTR, a novel inference-time defense that mitigates multimodal jailbreak attacks through optimizing the model's key-value (KV) caches. Rather than relying on curated safety-specific data or costly image-to-text conversion, we introduce a new formulation of the safety-relevant distributional shift induced by the visual modality. This formulation enables DTR to dynamically adjust visual token weights, minimizing the impact of adversarial visual inputs while preserving the model's general capabilities and inference efficiency. Extensive evaluation across diverse VLMs and attack benchmarks demonstrates that DTR outperforms existing defenses in both attack robustness and benign-task performance, marking the first successful application of KV cache optimization for safety enhancement in multimodal foundation models. The code for replicating DTR is available at: https://github.com/TanqiuJiang/DTR.
Authors: Kwanyoung Kim, Sanghyun Kim
Abstract: The choice of initial noise strongly affects quality and prompt alignment in video diffusion; different seeds for the same prompt can yield drastically different results. While recent methods use externally designed priors (e.g., frequency filtering or inter-frame smoothing), they often overlook internal model signals that indicate inherently preferable seeds. To address this, we propose ANSE (Active Noise Selection for Generation), a model-aware framework that selects high-quality seeds by quantifying attention-based uncertainty. At its core is BANSA (Bayesian Active Noise Selection via Attention), an acquisition function that measures entropy disagreement across multiple stochastic attention samples to estimate model confidence and consistency. For efficient inference-time deployment, we introduce a Bernoulli-masked approximation of BANSA that estimates scores from a single diffusion step and a subset of informative attention layers. Experiments across diverse text-to-video backbones demonstrate improved video quality and temporal coherence with marginal inference overhead, providing a principled and generalizable approach to noise selection in video diffusion. See our project page: https://anse-project.github.io/anse-project/.
Authors: Xueyang Li, Jiahao Li, Yu Song, Yunzhong Lou, Xiangdong Zhou
Abstract: The advent of Computer-Aided Design (CAD) generative modeling will significantly transform the design of industrial products. The recent research endeavor has extended into the realm of Large Language Models (LLMs). In contrast to fine-tuning methods, training-free approaches typically utilize the advanced closed-source LLMs, thereby offering enhanced flexibility and efficiency in the development of AI agents for generating CAD parametric models. However, the substantial cost and limitations of local deployment of the top-tier closed-source LLMs pose challenges in practical applications. The Seek-CAD is the pioneer exploration of locally deployed open-source inference LLM DeepSeek-R1 for CAD parametric model generation with a training-free methodology. This study is the first investigation to incorporate both visual and Chain-of-Thought (CoT) feedback within the self-refinement mechanism for generating CAD models. Specifically, the initial generated parametric CAD model is rendered into a sequence of step-wise perspective images, which are subsequently processed by a Vision Language Model (VLM) alongside the corresponding CoTs derived from DeepSeek-R1 to assess the CAD model generation. Then, the feedback is utilized by DeepSeek-R1 to refine the initial generated model for the next round of generation. Moreover, we present an innovative 3D CAD model dataset structured around the SSR (Sketch, Sketch-based feature, and Refinements) triple design paradigm. This dataset encompasses a wide range of CAD commands, thereby aligning effectively with industrial application requirements and proving suitable for the generation of LLMs. Extensive experiments validate the effectiveness of Seek-CAD under various metrics.
Authors: Xuweiyi Chen, Wentao Zhou, Aruni RoyChowdhury, Zezhou Cheng
Abstract: While massively scaling both data and models have become central in NLP and 2D vision, their benefits for 3D point cloud understanding remain limited. We study the initial step of scaling 3D point cloud understanding under a realistic regime: large-scale multi-dataset joint training for 3D semantic segmentation, with no dataset labels available at training or inference time. Point clouds arise from a wide range of sensors (e.g., depth cameras, LiDAR) and scenes (\eg, indoor, outdoor), yielding heterogeneous scanning patterns, sampling densities, and semantic biases; naively mixing such datasets degrades standard models. Therefore, we introduce Point-MoE, a Mixture-of-Experts design that expands model capacity through sparsely activated expert MLPs and a lightweight top-$k$ router, allowing tokens to select specialized experts without requiring dataset supervision. Trained jointly on a diverse mix of indoor and outdoor datasets, and evaluated on seen datasets as well as in zero-shot settings, Point-MoE outperforms prior methods without using dataset labels for either training or inference. This outlines a scalable path for 3D perception: letting the model discover structure in heterogeneous 3D data rather than imposing it via manual curation or dataset-specific heuristics.
Authors: Xingtong Ge, Xin Zhang, Tongda Xu, Yi Zhang, Xinjie Zhang, Yan Wang, Jun Zhang
Abstract: The Distribution Matching Distillation (DMD) has been successfully applied to text-to-image diffusion models such as Stable Diffusion (SD) 1.5. However, vanilla DMD suffers from convergence difficulties on large-scale flow-based text-to-image models, such as SD 3.5 and FLUX. In this paper, we first analyze the issues when applying vanilla DMD on large-scale models. Then, to overcome the scalability challenge, we propose implicit distribution alignment (IDA) to constrain the divergence between the generator and the fake distribution. Furthermore, we propose intra-segment guidance (ISG) to relocate the timestep denoising importance from the teacher model. With IDA alone, DMD converges for SD 3.5; employing both IDA and ISG, DMD converges for SD 3.5 and FLUX.1 dev. Together with a scaled VFM-based discriminator, our final model, dubbed \textbf{SenseFlow}, achieves superior performance in distillation for both diffusion based text-to-image models such as SDXL, and flow-matching models such as SD 3.5 Large and FLUX.1 dev. The source code is available at \href{https://github.com/XingtongGe/SenseFlow}{https://github.com/XingtongGe/SenseFlow}
URLs: https://github.com/XingtongGe/SenseFlow, https://github.com/XingtongGe/SenseFlow
Authors: Honglu Zhang, Zhiqin Fang, Ningning Zhao, Saihui Hou, Long Ma, Renwang Pei, Zhaofeng He
Abstract: Face Anti-Spoofing (FAS) typically depends on a single visual modality when defending against presentation attacks such as print attacks, screen replays, and 3D masks, resulting in limited generalization across devices, environments, and attack types. Meanwhile, Multimodal Large Language Models (MLLMs) have recently achieved breakthroughs in image-text understanding and semantic reasoning, suggesting that integrating visual and linguistic co-inference into FAS can substantially improve both robustness and interpretability. However, the lack of a high-quality vision-language multimodal dataset has been a critical bottleneck. To address this, we introduce FaceCoT (Face Chain-of-Thought), the first large-scale Visual Question Answering (VQA) dataset tailored for FAS. FaceCoT covers 14 spoofing attack types and enriches model learning with high-quality CoT VQA annotations. Meanwhile, we develop a caption model refined via reinforcement learning to expand the dataset and enhance annotation quality. Furthermore, we introduce a CoT-Enhanced Progressive Learning (CEPL) strategy to better leverage the CoT data and boost model performance on FAS tasks. Extensive experiments demonstrate that models trained with FaceCoT and CEPL outperform state-of-the-art methods on multiple benchmark datasets.
Authors: Mengdi Jia, Zekun Qi, Shaochen Zhang, Wenyao Zhang, Xinqiang Yu, Jiawei He, He Wang, Li Yi
Abstract: Spatial reasoning is a key aspect of cognitive psychology and remains a bottleneck for current vision-language models (VLMs). While extensive research has aimed to evaluate or improve VLMs' understanding of basic spatial relations, such as distinguishing left from right, near from far, and object counting, these tasks cover only the most elementary layer of spatial reasoning and are largely approaching saturation in the latest reasoning models. In this work, we introduce OmniSpatial, a comprehensive and challenging benchmark for spatial reasoning, grounded in cognitive psychology. OmniSpatial covers four major categories: dynamic reasoning, complex spatial logic, spatial interaction, and perspective-taking, with 50 fine-grained subcategories. Through careful manual annotation, we construct over 8.4K question-answer pairs. Extensive experiments show that both open- and closed-source VLMs exhibit significant limitations in comprehensive spatial reasoning. We also explore two strategies-PointGraph (explicit scene graph cues) and SpatialCoT (novel-view chain-of-thought)-to bolster spatial reasoning.
Authors: Mufhumudzi Muthivhi, Jiahao Huo, Fredrik Gustafsson, Terence L. van Zyl
Abstract: Mitigating human-wildlife conflict seeks to resolve unwanted encounters between these parties. Computer Vision provides a solution to identifying individuals that might escalate into conflict, such as members of the Big Five African animals. However, environments often contain several varied species. The current state-of-the-art animal classification models are trained under a closed-world assumption. They almost always remain overconfident in their predictions even when presented with unknown classes. This study investigates out-of-distribution (OOD) detection of wildlife, specifically the Big Five. To this end, we select a parametric Nearest Class Mean (NCM) and a non-parametric contrastive learning approach as baselines to take advantage of pretrained and projected features from popular classification encoders. Moreover, we compare our baselines to various common OOD methods in the literature. The results show feature-based methods reflect stronger generalisation capability across varying classification thresholds. Specifically, NCM with ImageNet pre-trained features achieves a 2%, 4% and 22% improvement on AUPR-IN, AUPR-OUT and AUTC over the best OOD methods, respectively. The code can be found here https://github.com/pxpana/BIG5OOD
Authors: Yukang Feng, Jianwen Sun, Chuanhao Li, Zizhen Li, Jiaxin Ai, Fanrui Zhang, Yifan Chang, Sizhuo Zhou, Shenglin Zhang, Yu Dai, Kaipeng Zhang
Abstract: Recent advancements in Large Multimodal Models (LMMs) have significantly improved multimodal understanding and generation. However, these models still struggle to generate tightly interleaved image-text outputs, primarily due to the limited scale, quality, and instructional richness of current training datasets. To address this, we introduce InterSyn, a dataset that features: (1) large scale, comprising 1.8M multimodal samples; (2) high quality, supported by our proposed Self-Evaluation with Iterative Refinement (SEIR) method for rigorous automated quality refinement; (3) rich instructional diversity, ensured through diverse well-designed question templates, based on human preferences and covering a 3500-topic hierarchy. These characteristics make InterSyn particularly well-suited for training LMMs in interactive image-text generation capabilities. To evaluate the capabilities, we propose SynJudge, a reliable automatic evaluator that aligns closely with human judge and outputs four interpretable scores: Text Content Completeness (TCC), Image Content Completeness (ICC), Image Quality (IQ), and Image-Text Synergy (ITS). These scores are complementary, covering both content and quality as well as cross-modal interaction, thereby forming a comprehensive evaluation framework. Experimental results on InterSyn subsets of up to 200K samples show that 25K-50K already yield substantial improvements, while scaling to 100K/200K brings further gains in TCC, ICC, and especially ITS, highlighting InterSyn's: (1) scalability, as performance consistently improves with more data; (2) efficiency, as significant gains are achievable even with smaller subsets, making it accessible to researchers with varying computational resources.
Authors: Haowen Wang, Xiaoping Yuan, Zhao Jin, Zhen Zhao, Zhengping Che, Yousong Xue, Jin Tian, Yakun Huang, Jian Tang
Abstract: Articulated objects are ubiquitous and important in robotics, AR/VR, and digital twins. Most self-supervised methods for articulated object modeling reconstruct discrete interaction states and relate them via cross-state geometric consistency, yielding representational fragmentation and drift that hinder smooth control of articulated configurations. We introduce PD$^{2}$GS, a novel framework that learns a shared canonical Gaussian field and models the arbitrary interaction state as its continuous deformation, jointly encoding geometry and kinematics. By associating each interaction state with a latent code and refining part boundaries using generic vision priors, PD$^{2}$GS enables accurate and reliable part-level decoupling while enforcing mutual exclusivity between parts and preserving scene-level coherence. This unified formulation supports part-aware reconstruction, fine-grained continuous control, and accurate kinematic modeling, all without manual supervision. To assess realism and generalization, we release RS-Art, a real-to-sim RGB-D dataset aligned with reverse-engineered 3D models, supporting real-world evaluation. Extensive experiments demonstrate that PD$^{2}$GS surpasses prior methods in geometric and kinematic accuracy, and in consistency under continuous control, both on synthetic and real data.
Authors: Christos Ziakas, Alessandra Russo
Abstract: Vision-Language Models (VLMs) show promise as zero-shot goal-conditioned value functions, but their frozen pre-trained representations limit generalization and temporal reasoning. We introduce VITA, a zero-shot value function learning method that enhances both capabilities via test-time adaptation. At inference, a lightweight adaptation module is updated via a gradient step on a meta-learned self-supervised loss, such that each test-time update improves value estimation. By updating sequentially over a trajectory, VITA encodes history into its parameters, addressing the temporal reasoning limitations. To mitigate shortcut learning, we propose a dissimilarity-based sampling strategy that selects semantically diverse segments of the trajectory during training. In real-world robotic manipulation tasks, VITA generalizes from a single training environment to diverse out-of-distribution tasks, environments, and embodiments, outperforming the state-of-the-art zero-shot method using autoregressive VLMs. Furthermore, we demonstrate that VITA's zero-shot value estimates can be utilized for reward shaping in offline reinforcement learning, resulting in multi-task policies on the Meta-World benchmark that exceed the performance of those trained with the simulation's fuzzy-logic dense rewards. Project website: https://chziakas.github.io/vita/.
Authors: Leigang Qu, Feng Cheng, Ziyan Yang, Qi Zhao, Shanchuan Lin, Yichun Shi, Yicong Li, Wenjie Wang, Tat-Seng Chua, Lu Jiang
Abstract: In-context image editing aims to modify images based on a contextual sequence comprising text and previously generated images. Existing methods typically depend on task-specific pipelines and expert models (e.g., segmentation and inpainting) to curate training data. In this work, we explore whether an in-context image editing model can be learned directly from videos. We introduce a scalable approach to annotate videos as interleaved multimodal sequences. To effectively learn from this data, we design a block-causal diffusion transformer trained on three proxy tasks: next-image prediction, current segmentation prediction, and next-segmentation prediction. Additionally, we propose a novel multi-turn image editing benchmark to advance research in this area. Extensive experiments demonstrate that our model exhibits strong in-context image editing capabilities and achieves state-of-the-art results on two multi-turn image editing benchmarks. Despite being trained exclusively on videos, our model also shows promising abilities in multi-concept composition, story generation, and chain-of-editing applications.
Authors: Rui Xu, Yunke Wang, Yong Luo, Bo Du
Abstract: Large Vision-Language Models (LVLMs) encode visual inputs as dense sequences of patch-level tokens to capture fine-grained semantics. These visual tokens often outnumber their textual counterparts by a large margin, leading to substantial computational overhead and limiting the scalability of LVLMs in practice. Previous efforts have explored visual token reduction either prior to or within the large language models (LLMs). However, most in-LLM reduction approaches rely on text-conditioned interactions, implicitly assuming that textual tokens can reliably capture the importance of visual tokens. In this work, we revisit this assumption and reveal causal, semantic, and spatial forms of cross-modal misalignment. These misalignments undermine the effectiveness of text-guided visual token reduction. To address this, we introduce VisionDrop, a training-free, visual-only pruning framework that selects informative visual tokens based on intra-modal (visual-to-visual) attention, without relying on textual signals. To further suppress redundancy throughout the model hierarchy, we treat the visual encoder and the LLM as a unified system and design a progressive pruning pipeline. Our method performs dominant token selection and lightweight contextual merging at multiple stages, enabling fine-grained visual information to be retained even under aggressive token budgets. Extensive experiments across diverse benchmarks show that VisionDrop achieves consistent improvements over existing approaches, despite requiring no additional training or complex modifications. Notably, when integrated with LLaVA-NeXT-7B, VisionDrop achieves a 2.7x reduction in inference latency and 6x in FLOPs, while retaining 95.71% of the original performance.
Authors: JaeHyuck Choi, MinJun Kim, Je Hyeong Hong
Abstract: Few-shot anomaly generation is a key challenge in industrial quality control. Although diffusion models are promising, existing methods struggle: global prompt-guided approaches corrupt normal regions, and existing inpainting-based methods often lack the in-distribution diversity essential for robust downstream models. We propose MAGIC, a fine-tuned inpainting framework that generates high-fidelity anomalies that strictly adhere to the mask while maximizing this diversity. MAGIC introduces three complementary components: (i) Gaussian prompt perturbation, which prevents model overfitting in the few-shot setting by learning and sampling from a smooth manifold of realistic anomalies, (ii) spatially adaptive guidance that applies distinct guidance strengths to the anomaly and background regions, and (iii) context-aware mask alignment to relocate masks for plausible placement within the host object. Under consistent identical evaluation protocol, MAGIC outperforms state-of-the-art methods on diverse anomaly datasets in downstream tasks
Authors: Gaochao Song, Zibo Zhao, Haohan Weng, Jingbo Zeng, Rongfei Jia, Shenghua Gao
Abstract: Existing auto-regressive mesh generation approaches suffer from ineffective topology preservation, which is crucial for practical applications. This limitation stems from previous mesh tokenization methods treating meshes as simple collections of equivalent triangles, lacking awareness of the overall topological structure during generation. To address this issue, we propose a novel mesh tokenization algorithm that provides a canonical topological framework through vertex layering and ordering, ensuring critical geometric properties including manifoldness, watertightness, face normal consistency, and part awareness in the generated meshes. Measured by Compression Ratio and Bits-per-face, we also achieved state-of-the-art compression efficiency. Furthermore, we introduce an online non-manifold data processing algorithm and a training resampling strategy to expand the scale of trainable dataset and avoid costly manual data curation. Experimental results demonstrate the effectiveness of our approach, showcasing not only intricate mesh generation but also significantly improved geometric integrity.
Authors: Jiuhong Xiao, Yang Zhou, Giuseppe Loianno
Abstract: Deep learning methods for Visual Place Recognition (VPR) have advanced significantly, largely driven by large-scale datasets. However, most existing approaches are trained on a single dataset, which can introduce dataset-specific inductive biases and limit model generalization. While multi-dataset joint training offers a promising solution for developing universal VPR models, divergences among training datasets can saturate the limited information capacity in feature aggregation layers, leading to suboptimal performance. To address these challenges, we propose Query-based Adaptive Aggregation (QAA), a novel feature aggregation technique that leverages learned queries as reference codebooks to effectively enhance information capacity without significant computational or parameter complexity. We show that computing the Cross-query Similarity (CS) between query-level image features and reference codebooks provides a simple yet effective way to generate robust descriptors. Our results demonstrate that QAA outperforms state-of-the-art models, achieving balanced generalization across diverse datasets while maintaining peak performance comparable to dataset-specific models. Ablation studies further explore QAA's mechanisms and scalability. Visualizations reveal that the learned queries exhibit diverse attention patterns across datasets. Project page: http://xjh19971.github.io/QAA.
Authors: Rang Meng, Yan Wang, Weipeng Wu, Ruobing Zheng, Yuming Li, Chenguang Ma
Abstract: Recent work on human animation usually incorporates large-scale video models, thereby achieving more vivid performance. However, the practical use of such methods is hindered by the slow inference speed and high computational demands. Moreover, traditional work typically employs separate models for each animation task, increasing costs in multi-task scenarios and worsening the dilemma. To address these limitations, we introduce EchoMimicV3, an efficient framework that unifies multi-task and multi-modal human animation. At the core of EchoMimicV3 lies a threefold design: a Soup-of-Tasks paradigm, a Soup-of-Modals paradigm, and a novel training and inference strategy. The Soup-of-Tasks leverages multi-task mask inputs and a counter-intuitive task allocation strategy to achieve multi-task gains without multi-model pains. Meanwhile, the Soup-of-Modals introduces a Coupled-Decoupled Multi-Modal Cross Attention module to inject multi-modal conditions, complemented by a Multi-Modal Timestep Phase-aware Dynamical Allocation mechanism to modulate multi-modal mixtures. Besides, we propose Negative Direct Preference Optimization, Phase-aware Negative Classifier-Free Guidance (CFG), and Long Video CFG, which ensure stable training and inference. Extensive experiments and analyses demonstrate that EchoMimicV3, with a minimal model size of 1.3 billion parameters, achieves competitive performance in both quantitative and qualitative evaluations.
Authors: Yonghyun Park, Chieh-Hsin Lai, Satoshi Hayakawa, Yuhta Takida, Naoki Murata, Wei-Hsiang Liao, Woosung Choi, Kin Wai Cheuk, Junghyun Koo, Yuki Mitsufuji
Abstract: While diffusion models excel at image generation, their growing adoption raises critical concerns about copyright issues and model transparency. Existing attribution methods identify training examples influencing an entire image, but fall short in isolating contributions to specific elements, such as styles or objects, that are of primary concern to stakeholders. To address this gap, we introduce concept-level attribution through a novel method called Concept-TRAK, which extends influence functions with a key innovation: specialized training and utility loss functions designed to isolate concept-specific influences rather than overall reconstruction quality. We evaluate Concept-TRAK on novel concept attribution benchmarks using Synthetic and CelebA-HQ datasets, as well as the established AbC benchmark, showing substantial improvements over prior methods in concept-level attribution scenarios. We further demonstrate its versatility on real-world text-to-image generation with compositional and multi-concept prompts.
Authors: Zhengqing Wang, Yuefan Wu, Jiacheng Chen, Fuyang Zhang, Yasutaka Furukawa
Abstract: This paper proposes a neural rendering approach that represents a scene as "compressed light-field tokens (CLiFTs)", retaining rich appearance and geometric information of a scene. CLiFT enables compute-efficient rendering by compressed tokens, while being capable of changing the number of tokens to represent a scene or render a novel view with one trained network. Concretely, given a set of images, multi-view encoder tokenizes the images with the camera poses. Latent-space K-means selects a reduced set of rays as cluster centroids using the tokens. The multi-view ``condenser'' compresses the information of all the tokens into the centroid tokens to construct CLiFTs. At test time, given a target view and a compute budget (i.e., the number of CLiFTs), the system collects the specified number of nearby tokens and synthesizes a novel view using a compute-adaptive renderer. Extensive experiments on RealEstate10K and DL3DV datasets quantitatively and qualitatively validate our approach, achieving significant data reduction with comparable rendering quality and the highest overall rendering score, while providing trade-offs of data size, rendering quality, and rendering speed.
Authors: Zhixiong Zhang, Shuangrui Ding, Xiaoyi Dong, Songxin He, Jianfan Lin, Junsong Tang, Yuhang Zang, Yuhang Cao, Dahua Lin, Jiaqi Wang
Abstract: We propose Segment Concept (SeC), a concept-driven video object segmentation (VOS) framework that shifts from conventional feature matching to the progressive construction and utilization of high-level, object-centric representations. SeC employs Large Vision-Language Models (LVLMs) to integrate visual cues across diverse frames, constructing robust conceptual priors. To balance semantic reasoning with computational overhead, SeC forwards the LVLMs only when a new scene appears, injecting concept-level features at those points. To rigorously assess VOS methods in scenarios demanding high-level conceptual reasoning and robust semantic understanding, we introduce the Semantic Complex Scenarios Video Object Segmentation benchmark (SeCVOS). SeCVOS comprises 160 manually annotated multi-scenario videos designed to challenge models with substantial appearance variations and dynamic scene transformations. Empirical evaluations demonstrate that SeC substantially outperforms state-of-the-art approaches, including SAM 2 and its advanced variants, on both SeCVOS and standard VOS benchmarks. In particular, SeC achieves an 11.8-point improvement over SAM 2.1 on SeCVOS, establishing a new state-of-the-art in concept-aware VOS.
Authors: Zihan Wang, Jeff Tan, Tarasha Khurana, Neehar Peri, Deva Ramanan
Abstract: We address the problem of dynamic scene reconstruction from sparse-view videos. Prior work often requires dense multi-view captures with hundreds of calibrated cameras (e.g. Panoptic Studio). Such multi-view setups are prohibitively expensive to build and cannot capture diverse scenes in-the-wild. In contrast, we aim to reconstruct dynamic human behaviors, such as repairing a bike or dancing, from a small set of sparse-view cameras with complete scene coverage (e.g. four equidistant inward-facing static cameras). We find that dense multi-view reconstruction methods struggle to adapt to this sparse-view setup due to limited overlap between viewpoints. To address these limitations, we carefully align independent monocular reconstructions of each camera to produce time- and view-consistent dynamic scene reconstructions. Extensive experiments on PanopticStudio and Ego-Exo4D demonstrate that our method achieves higher quality reconstructions than prior art, particularly when rendering novel views. Code, data, and data-processing scripts are available on https://github.com/Z1hanW/MonoFusion.
Authors: Hao Si, Xiao Wang, Fan Zhang, Xiaoya Zhou, Dengdi Sun, Wanli Lyu, Qingquan Yang, Jin Tang
Abstract: Multivariate time series analysis has long been one of the key research topics in the field of artificial intelligence. However, analyzing complex time series data remains a challenging and unresolved problem due to its high dimensionality, dynamic nature, and complex interactions among variables. Inspired by the strong structural modeling capability of hypergraphs, this paper proposes a novel hypergraph-based time series Transformer backbone network, termed HGTS-Former, to address the multivariate coupling in time series data. Specifically, given the multivariate time series signal, we first normalize and embed each patch into tokens. Then, we adopt the multi-head self-attention to enhance the temporal representation of each patch. The hierarchical hypergraphs are constructed to aggregate the temporal patterns within each channel and fine-grained relations between different variables. After that, we convert the hyperedge into node features through the EdgeToNode module and adopt the feed-forward network to further enhance the output features. Extensive experiments on multiple representative time series analysis tasks and public datasets fully validated the effectiveness of our proposed HGTS-Former. Moreover, we present EAST-ELM640, a large-scale time series dataset for Edge-Localized Mode (ELM) recognition in nuclear fusion, on which we achieve state-of-the-art performance. The source code will be released on https://github.com/Event-AHU/Time_Series_Analysis
Authors: Minh-Hai Nguyen, Edouard Pauwels, Pierre Weiss
Abstract: Maximum A Posteriori (MAP) estimation is a cornerstone framework for blind inverse problems, where an image and a forward operator are jointly estimated as the maximizers of a posterior distribution. In this paper, we analyze the recovery guarantees of MAP-based methods by adopting a Morse-Bott framework. We model the image prior potential as a Morse-Bott function, where natural images are modeled as residing locally on a critical submanifold. This means that while the potential is locally flat along the natural directions of the image manifold, it is strictly convex in the directions normal to it. We demonstrate that this Morse-Bott hypothesis aligns with the structural properties of state-of-the-art learned priors, a finding we validate through an experimental analysis of the potential landscape and its Hessian spectrum. Our theoretical results show that, in a neighborhood of the ground-truth image and operator, the posterior admits local minimizers that are stable both with respect to initialization (gradient steps converge to the same minimizer) and to small noise perturbations (solutions vary smoothly). This local stability explains the empirical success of well-designed gradient-based optimization in these settings. However, we also demonstrate that this stability is a local property: the blurry trap, well-known for sparse priors in blind deconvolution, persists even with state-of-the-art learned priors. Our findings demonstrate that the failure of MAP in blind deconvolution is not a limitation of prior quality, but an intrinsic characteristic of the landscape. We conclude that successful recovery of posterior maximization depends on strategic initialization within the basin of favorable local minima, and we validate this with numerical experiments on both synthetic and real-world data.
Authors: Bo Zhang, Yifan Zhang, Shuo Yan, Yu Bai, Zheng Zhang, Wu Liu, Wendong Wang, Yongdong Zhang
Abstract: Accurate 3D medical image segmentation requires a delicate balance between fine-grained local details and global contextual understanding. While spatial-domain models often struggle with long-range dependencies, existing frequency-based approaches frequently overlook intrinsic spectral properties such as Hermitian symmetry, leading to suboptimal feature integration. In this paper, we propose SSFMamba, a Mamba based Symmetry-driven Spatial-Frequency fusion framework tailored for 3D medical imaging. Our architecture employs a complementary dual-branch design: the spatial branch preserves intricate anatomical textures, while the frequency branch captures global contextual dependencies in the frequency domain. A core innovation is the 3D Multi-Directional Scanning Mechanism (MDSM), which integrates Hermitian symmetry with the causal nature of State Space Models (SSMs) to enable direction-aware global modeling. Crucially, by shifting the modeling focus to frequency-domain spectral components, SSFMamba captures the underlying structural characteristics of anatomical tissues. This leads to a highly adaptable framework that excels in both MRI and CT applications, regardless of the significant variations in intensity distributions. Extensive evaluations on the BraTS2020, BraTS2023, and BTCV datasets demonstrate that SSFMamba consistently outperforms state-of-the-art methods. Notably, our approach achieves exceptional performance on low-contrast organs such as the pancreas (81.97% Dice), underscoring its potential as a unified and physically consistent perception framework for diverse 3D clinical applications.
Authors: Panagiotis Alimisis, Christos Diou
Abstract: Causal representation learning has attracted significant research interest during the past few years, as a means for improving model generalization and robustness. Causal representations of interventional image pairs (also called ``actionable counterfactuals'' in the literature), have the property that only variables corresponding to scene elements affected by the intervention / action are changed between the start state and the end state. While most work in this area has focused on identifying and representing the variables of the scene under a causal model, fewer efforts have focused on representations of the interventions themselves. In this work, we show that an effective strategy for improving out of distribution (OOD) robustness is to focus on the representation of actionable counterfactuals in the latent space. Specifically, we propose that an intervention can be represented by a Causal Delta Embedding that is invariant to the visual scene and sparse in terms of the causal variables it affects. Leveraging this insight, we propose a method for learning causal representations from image pairs, without any additional supervision. Experiments in the Causal Triplet challenge demonstrate that Causal Delta Embeddings are highly effective in OOD settings, significantly exceeding baseline performance in both synthetic and real-world benchmarks.
Authors: Young D. Kwon, Rui Li, Sijia Li, Da Li, Sourav Bhattacharya, Stylianos I. Venieris
Abstract: State-of-the-art text-to-image diffusion models (DMs) achieve remarkable quality, yet their massive parameter scale (8-11B) poses significant challenges for inferences on resource-constrained devices. In this paper, we present HierarchicalPrune, a novel compression framework grounded in a key observation: DM blocks exhibit distinct functional hierarchies, where early blocks establish semantic structures while later blocks handle texture refinements. HierarchicalPrune synergistically combines three techniques: (1) Hierarchical Position Pruning, which identifies and removes less essential later blocks based on position hierarchy; (2) Positional Weight Preservation, which systematically protects early model portions that are essential for semantic structural integrity; and (3) Sensitivity-Guided Distillation, which adjusts knowledge-transfer intensity based on our discovery of block-wise sensitivity variations. As a result, our framework brings billion-scale diffusion models into a range more suitable for on-device inference, while preserving the quality of the output images. Specifically, combined with INT4 weight quantisation, HierarchicalPrune achieves 77.5-80.4% memory footprint reduction (e.g., from 15.8 GB to 3.2 GB) and 27.9-38.0% latency reduction, measured on server and consumer grade GPUs, with the minimum drop of 2.6% in GenEval score and 7% in HPSv2 score compared to the original model. Finally, our comprehensive user study with 85 participants demonstrates that HierarchicalPrune maintains perceptual quality comparable to the original model while significantly outperforming prior works.
Authors: Luozheng Qin, Jia Gong, Yuqing Sun, Tianjiao Li, Mengping Yang, Xiaomeng Yang, Chao Qu, Zhiyu Tan, Hao Li
Abstract: Chain-of-Thought (CoT) reasoning has been widely adopted to enhance Large Language Models (LLMs) by decomposing complex tasks into simpler, sequential subtasks. However, extending CoT to vision-language reasoning tasks remains challenging, as it often requires interpreting transitions of visual states to support reasoning. Existing methods often struggle with this due to limited capacity of modeling visual state transitions or incoherent visual trajectories caused by fragmented architectures. To overcome these limitations, we propose Uni-CoT, a Unified Chain-of-Thought framework that enables coherent and grounded multimodal reasoning within a single unified model. The key idea is to leverage a model capable of both image understanding and generation to reason over visual content and model evolving visual states. However, empowering a unified model to achieve that is non-trivial, given the high computational cost and the burden of training. To address this, Uni-CoT introduces a novel two-level reasoning paradigm: A Macro-Level CoT for high-level task planning and A Micro-Level CoT for subtask execution. This design significantly reduces the computational overhead. Furthermore, we introduce a structured training paradigm that combines interleaved image-text supervision for macro-level CoT with multi-task objectives for micro-level CoT. Together, these innovations allow Uni-CoT to perform scalable and coherent multi-modal reasoning. Furthermore, thanks to our design, all experiments can be efficiently completed using only 8 A100 GPUs with 80GB VRAM each. Experimental results on reasoning-driven image generation benchmark (WISE) and editing benchmarks (RISE and KRIS) indicates that Uni-CoT demonstrates SOTA performance and strong generalization, establishing Uni-CoT as a promising solution for multi-modal reasoning. Project Page and Code: https://sais-fuxi.github.io/projects/uni-cot/
Authors: Jingyu Li, Bozhou Zhang, Xin Jin, Jiankang Deng, Xiatian Zhu, Li Zhang
Abstract: Autonomous driving requires rich contextual comprehension and precise predictive reasoning to navigate dynamic and complex environments safely. Vision-Language Models (VLMs) and Driving World Models (DWMs) have independently emerged as powerful recipes addressing different aspects of this challenge. VLMs provide interpretability and robust action prediction through their ability to understand multi-modal context, while DWMs excel in generating detailed and plausible future driving scenarios essential for proactive planning. Integrating VLMs with DWMs is an intuitive, promising, yet understudied strategy to exploit the complementary strengths of accurate behavioral prediction and realistic scene generation. Nevertheless, this integration presents notable challenges, particularly in effectively connecting action-level decisions with high-fidelity pixel-level predictions and maintaining computational efficiency. In this paper, we propose ImagiDrive, a novel end-to-end autonomous driving framework that integrates a VLM-based driving agent with a DWM-based scene imaginer to form a unified imagination-and-planning loop. The driving agent predicts initial driving trajectories based on multi-modal inputs, guiding the scene imaginer to generate corresponding future scenarios. These imagined scenarios are subsequently utilized to iteratively refine the driving agent's planning decisions. To address efficiency and predictive accuracy challenges inherent in this integration, we introduce an early stopping mechanism and a trajectory selection strategy. Extensive experimental validation on the nuScenes and NAVSIM datasets demonstrates the robustness and superiority of ImagiDrive over previous alternatives under both open-loop and closed-loop conditions.
Authors: Xiaoxue Wu, Bingjie Gao, Yu Qiao, Yaohui Wang, Xinyuan Chen
Abstract: Despite significant advances in video synthesis, research into multi-shot video generation remains in its infancy. Even with scaled-up models and massive datasets, the shot transition capabilities remain rudimentary and unstable, largely confining generated videos to single-shot sequences. In this work, we introduce CineTrans, a novel framework for generating coherent multi-shot videos with cinematic, film-style transitions. To facilitate insights into the film editing style, we construct a multi-shot video-text dataset Cine250K with detailed shot annotations. Furthermore, our analysis of existing video diffusion models uncovers a correspondence between attention maps in the diffusion model and shot boundaries, which we leverage to design a mask-based control mechanism that enables transitions at arbitrary positions and transfers effectively in a training-free setting. After fine-tuning on our dataset with the mask mechanism, CineTrans produces cinematic multi-shot sequences while adhering to the film editing style, avoiding unstable transitions or naive concatenations. Finally, we propose specialized evaluation metrics for transition control, temporal consistency and overall quality, and demonstrate through extensive experiments that CineTrans significantly outperforms existing baselines across all criteria.
Authors: Daoze Zhang, Chenghan Fu, Zhanheng Nie, Jianyu Liu, Wanxian Guan, Yuan Gao, Jun Song, Pengjie Wang, Jian Xu, Bo Zheng
Abstract: With the rapid advancement of e-commerce, exploring general representations rather than task-specific ones has attracted increasing research attention. For product understanding, although existing discriminative dual-flow architectures drive progress in this field, they inherently struggle to model the many-to-one alignment between multiple images and texts of products. Therefore, we argue that generative Multimodal Large Language Models (MLLMs) hold significant potential for improving product representation learning. Nevertheless, achieving this goal still remains non-trivial due to several key challenges: the lack of multimodal and aspect-aware modeling modules in typical LLMs; the common presence of background noise in product images; and the absence of a standard benchmark for evaluation. To address these issues, we propose the first generative MLLM-based model named MOON for product representation learning. Our method (1) employs a guided Mixture-of-Experts (MoE) module for targeted modeling of multimodal and aspect-specific product content; (2) effectively detects core semantic regions in product images to mitigate the distraction and interference caused by background noise; and (3) introduces the specialized negative sampling strategy to increase the difficulty and diversity of negative samples. In addition, we release a large-scale multimodal benchmark MBE for various product understanding tasks. Experimentally, our model demonstrates competitive zero-shot performance on both our benchmark and the public dataset, showcasing strong generalization across various downstream tasks, including cross-modal retrieval, product classification, and attribute prediction. Furthermore, the case study and visualization illustrate the effectiveness of MOON for product understanding. The data of our MBE benchmark is given in https://huggingface.co/datasets/Daoze/MM-Bench-E-Commerce.
URLs: https://huggingface.co/datasets/Daoze/MM-Bench-E-Commerce.
Authors: Yikai Wang, Zhouxia Wang, Zhonghua Wu, Qingyi Tao, Kang Liao, Chen Change Loy
Abstract: We propose a novel approach to image generation by decomposing an image into a structured sequence, where each element in the sequence shares the same spatial resolution but differs in the number of unique tokens used, capturing different level of visual granularity. Image generation is carried out through our newly introduced Next Visual Granularity (NVG) generation framework, which generates a visual granularity sequence beginning from an empty image and progressively refines it, from global layout to fine details, in a structured manner. This iterative process encodes a hierarchical, layered representation that offers fine-grained control over the generation process across multiple granularity levels. We train a series of NVG models for class-conditional image generation on the ImageNet dataset and observe clear scaling behavior. Compared to the VAR series, NVG consistently outperforms it in terms of FID scores (3.30 $\rightarrow$ 3.03, 2.57 $\rightarrow$ 2.44, 2.09 $\rightarrow$ 2.06). We also conduct extensive analysis to showcase the capability and potential of the NVG framework. Our code and models are released at https://yikai-wang.github.io/nvg.
Authors: Chubin Chen, Jiashu Zhu, Xiaokun Feng, Nisha Huang, Chen Zhu, Meiqi Wu, Fangyuan Mao, Jiahong Wu, Xiangxiang Chu, Xiu Li
Abstract: Classifier-free Guidance (CFG) is a widely used technique in modern diffusion models for enhancing sample quality and prompt adherence. However, through an empirical analysis on Gaussian mixture modeling with a closed-form solution, we observe a discrepancy between the suboptimal results produced by CFG and the ground truth. The model's excessive reliance on these suboptimal predictions often leads to semantic incoherence and low-quality outputs. To address this issue, we first empirically demonstrate that the model's suboptimal predictions can be effectively refined using sub-networks of the model itself. Building on this insight, we propose S$^2$-Guidance, a novel method that leverages stochastic block-dropping during the forward process to construct stochastic sub-networks, effectively guiding the model away from potential low-quality predictions and toward high-quality outputs. Extensive qualitative and quantitative experiments on text-to-image and text-to-video generation tasks demonstrate that S$^2$-Guidance delivers superior performance, consistently surpassing CFG and other advanced guidance strategies. Our code will be released.
Authors: Yizhou Liu, Dingkang Yang, Zizhi Chen, Minghao Han, Xukun Zhang, Keliang Liu, Jingwei Wei, Lihua Zhang
Abstract: Reinforcement learning (RL) with rule-based reward functions has recently shown great promise in enhancing the reasoning depth and generalization ability of vision-language models (VLMs), while maintaining computational efficiency. In spite of these advances, its adoption in medical imaging remains limited. Current reinforcement fine-tuning (RFT) efforts in this field mainly focus on closed-ended visual question answering (VQA), restricting their applicability to realistic clinical reasoning. However, open-ended medical VQA better mirrors clinical diagnostic workflows but remains underexplored. Although several studies have attempted to bridge the two formats through semantically guided RL, model-driven semantic rewards often suffer from reward collapse, where responses with distinct semantics yield nearly identical scores. To overcome this limitation, we introduce Adaptive Reinforcement for Medical Reasoning (ARMed), a novel RL framework tailored for open-ended medical VQA. ARMed first injects domain expertise through supervised fine-tuning (SFT) on chain-of-thought annotations, followed by reinforcement optimization using textual correctness and adaptive semantic rewards to refine reasoning consistency and factual accuracy. Extensive experiments on six challenging medical VQA benchmarks demonstrate that ARMed substantially improves both accuracy and generalization. These findings underscore the importance of reward discriminability in medical RL and highlight the potential of adaptive semantic rewards for building robust, clinically reliable multimodal reasoning systems.
Authors: Minhao Xiong, Zichen Wen, Zhuangcheng Gu, Xuyang Liu, Rui Zhang, Hengrui Kang, Jiabing Yang, Junyuan Zhang, Weijia Li, Conghui He, Yafei Wang, Linfeng Zhang
Abstract: Vision-Language Models (VLMs) have emerged as a promising paradigm in autonomous driving (AD), providing a unified framework for perception and decision-making. However, their real-world deployment is hindered by significant computational overhead when processing high-resolution, multi-view images. This complexity stems from the massive number of visual tokens, which increases inference latency and memory consumption due to the quadratic complexity of self-attention. To address these challenges, we propose Prune2Drive, a plug-and-play visual token pruning framework for multi-view VLMs in AD. Prune2Drive introduces two core innovations: (i) a diversity-aware token selection mechanism that prioritizes semantic and spatial coverage across views, and (ii) a view-adaptive pruning controller that automatically learns optimal pruning ratios based on camera importance to downstream tasks. Unlike prior methods, Prune2Drive requires no model retraining or access to attention maps, ensuring compatibility with modern efficient attention implementations. Extensive experiments on the DriveLM and DriveLMM-o1 benchmarks demonstrate that Prune2Drive achieves significant speedups and memory savings with minimal performance impact. When retaining only 10% of visual tokens, our method achieves a 6.40x speedup in the prefilling phase and consumes only 13.4% of the original FLOPs, with a mere 3% average performance drop on the DriveLM benchmark. Code is available at: https://github.com/MinhaoXiong/Prune2Drive.git
Authors: Sankalp Pandey, Xuan Bac Nguyen, Nicholas Borys, Hugh Churchill, Khoa Luu
Abstract: Identifying quantum flakes is crucial for scalable quantum hardware; however, automated layer classification from optical microscopy remains challenging due to substantial appearance shifts across different materials. This paper proposes a new Continual-Learning Framework for Flake Layer Classification (CLIFF). To the best of our knowledge, this work represents the first systematic study of continual learning in two-dimensional (2D) materials. The proposed framework enables the model to distinguish materials and their physical and optical properties by freezing the backbone and base head, which are trained on a reference material. For each new material, it learns a material-specific prompt, embedding, and a delta head. A prompt pool and a cosine-similarity gate modulate features and compute material-specific corrections. Additionally, memory replay with knowledge distillation is incorporated. CLIFF achieves competitive accuracy with significantly lower forgetting than naive fine-tuning and a prompt-based baseline.
Authors: Yue Wu, Xuanhong Chen, Yufan Wu, Wen Li, Yuxi Lu, Kairui Feng
Abstract: Despite significant progress in 3D avatar reconstruction, it still faces challenges such as high time complexity, sensitivity to data quality, and low data utilization. We propose FastAvatar, a feedforward 3D avatar framework capable of flexibly leveraging diverse daily recordings (e.g., a single image, multi-view observations, or monocular video) to reconstruct a high-quality 3D Gaussian Splatting (3DGS) model within seconds, using only a single unified model. The core of FastAvatar is a Large Gaussian Reconstruction Transformer (LGRT) featuring three key designs: First, a 3DGS transformer aggregating multi-frame cues while injecting initial 3D prompt to predict the corresponding registered canonical 3DGS representations; Second, multi-granular guidance encoding (camera pose, expression coefficient, head pose) mitigating animation-induced misalignment for variable-length inputs; Third, incremental Gaussian aggregation via landmark tracking and sliced fusion losses. Integrating these features, FastAvatar enables incremental reconstruction, i.e., improving quality with more observations without wasting input data as in previous works. This yields a quality-speed-tunable paradigm for highly usable 3D avatar modeling. Extensive experiments show that FastAvatar has a higher quality and highly competitive speed compared to existing methods.
Authors: Yuquan Bi, Hongsong Wang, Xinli Shi, Zhipeng Gui, Jie Gui, Yuan Yan Tang
Abstract: Diffusion models have demonstrated strong capabilities in generating high-fidelity 3D human poses, yet their iterative nature and multi-hypothesis requirements incur substantial computational cost. In this paper, we propose an Efficient Diffusion-Based 3D Human Pose Estimation framework with a Hierarchical Temporal Pruning (HTP) strategy, which dynamically prunes redundant pose tokens across both frame and semantic levels while preserving critical motion dynamics. HTP operates in a staged, top-down manner: (1) Temporal Correlation-Enhanced Pruning (TCEP) identifies essential frames by analyzing inter-frame motion correlations through adaptive temporal graph construction; (2) Sparse-Focused Temporal MHSA (SFT MHSA) leverages the resulting frame-level sparsity to reduce attention computation, focusing on motion-relevant tokens; and (3) Mask-Guided Pose Token Pruner (MGPTP) performs fine-grained semantic pruning via clustering, retaining only the most informative pose tokens. Experiments on Human3.6M and MPI-INF-3DHP show that HTP reduces training MACs by 38.5\%, inference MACs by 56.8\%, and improves inference speed by an average of 81.1\% compared to prior diffusion-based methods, while achieving state-of-the-art performance.
Authors: Shan Wang, Maying Shen, Nadine Chang, Chuong Nguyen, Hongdong Li, Jose M. Alvarez
Abstract: Multimodal large language models achieve strong performance across diverse tasks but remain prone to hallucinations, where outputs are not grounded in visual inputs. This issue can be attributed to two main biases: text-visual bias, the overreliance on prompts and prior outputs, and co-occurrence bias, spurious correlations between frequently paired objects. We propose Gradient-based Influence-Aware Constrained Decoding (GACD), an inference-based method, that addresses both biases without auxiliary models, and is readily applicable to existing models without finetuning. The core of our approach is bias estimation, which uses first-order Taylor gradients to understand the contribution of individual tokens-visual features and text tokens-to the current output. Based on this analysis, GACD mitigates hallucinations through two components: (1) suppressing spurious visual features correlated with the output objects, and (2) rebalancing cross-modal contributions by strengthening visual features relative to text. Experiments across multiple benchmarks demonstrate that GACD effectively reduces hallucinations and improves the visual grounding of MLLM outputs.
Authors: Junhao Jia, Yifei Sun, Yunyou Liu, Cheng Yang, Changmiao Wang, Feiwei Qin, Yong Peng, Wenwen Min
Abstract: Functional magnetic resonance imaging (fMRI) is a powerful tool for probing brain function, yet reliable clinical diagnosis is hampered by low signal-to-noise ratios, inter-subject variability, and the limited frequency awareness of prevailing CNN- and Transformer-based models. Moreover, most fMRI datasets lack textual annotations that could contextualize regional activation and connectivity patterns. We introduce RTGMFF, a framework that unifies automatic ROI-level text generation with multimodal feature fusion for brain-disorder diagnosis. RTGMFF consists of three components: (i) ROI-driven fMRI text generation deterministically condenses each subject's activation, connectivity, age, and sex into reproducible text tokens; (ii) Hybrid frequency-spatial encoder fuses a hierarchical wavelet-mamba branch with a cross-scale Transformer encoder to capture frequency-domain structure alongside long-range spatial dependencies; and (iii) Adaptive semantic alignment module embeds the ROI token sequence and visual features in a shared space, using a regularized cosine-similarity loss to narrow the modality gap. Extensive experiments on the ADHD-200 and ABIDE benchmarks show that RTGMFF surpasses current methods in diagnostic accuracy, achieving notable gains in sensitivity, specificity, and area under the ROC curve. Code is available at https://github.com/BeistMedAI/RTGMFF.
Authors: Ouxiang Li, Yuan Wang, Xinting Hu, Huijuan Huang, Rui Chen, Jiarong Ou, Xin Tao, Pengfei Wan, Xiaojuan Qi, Fuli Feng
Abstract: Text-to-image (T2I) generation aims to synthesize images from textual prompts, which jointly specify what must be shown and imply what can be inferred, which thus correspond to two core capabilities: \textbf{\textit{composition}} and \textbf{\textit{reasoning}}. Despite recent advances of T2I models in both composition and reasoning, existing benchmarks remain limited in evaluation. They not only fail to provide comprehensive coverage across and within both capabilities, but also largely restrict evaluation to low scene density and simple one-to-one reasoning. To address these limitations, we propose \textbf{\textsc{T2I-CoReBench}}, a comprehensive and complex benchmark that evaluates both composition and reasoning capabilities of T2I models. To ensure comprehensiveness, we structure composition around scene graph elements (\textit{instance}, \textit{attribute}, and \textit{relation}) and reasoning around the philosophical framework of inference (\textit{deductive}, \textit{inductive}, and \textit{abductive}), formulating a 12-dimensional evaluation taxonomy. To increase complexity, driven by the inherent real-world complexities, we curate each prompt with higher compositional density for composition and greater reasoning intensity for reasoning. To facilitate fine-grained and reliable evaluation, we also pair each evaluation prompt with a checklist that specifies individual \textit{yes/no} questions to assess each intended element independently. In statistics, our benchmark comprises 1,080 challenging prompts and around 13,500 checklist questions. Experiments across 38 current T2I models reveal that their composition capability still remains limited in high compositional scenarios, while the reasoning capability lags even further behind as a critical bottleneck, with all models struggling to infer implicit elements from prompts.
Authors: Haowang Cui, Rui Chen, Jiaze Wang, Tao Guo, Zheng Qin
Abstract: The task of synthesizing novel views from a single image is highly ill-posed due to multiple explanations for unobserved areas. Most current methods tend to generate unseen regions from ambiguity priors and interpolation near input views, which often lead to severe distortions. To address this limitation, we propose a novel model dubbed as UniView, which can leverage reference images from a similar object to provide strong prior information during view synthesis. More specifically, we construct a retrieval and augmentation system and employ a multimodal large language model (MLLM) to assist in selecting reference images that meet our requirements. Additionally, a plug-and-play adapter module with multi-level isolation layers is introduced to dynamically generate reference features for the target views. Moreover, in order to preserve the details of an original input image, we design a decoupled triple attention mechanism, which can effectively align and integrate multi-branch features into the synthesis process. Extensive experiments have demonstrated that our UniView significantly improves novel view synthesis performance and outperforms state-of-the-art methods on the challenging datasets.
Authors: Haoming Lu
Abstract: Skin tone recognition and generation play important roles in model fairness, healthcare, and generative AI, yet they remain challenging due to the lack of comprehensive datasets and robust methodologies. Compared to other human image analysis tasks, state-of-the-art large multimodal models (LMMs) and image generation models struggle to recognize and synthesize skin tones accurately. To address this, we introduce TrueSkin, a dataset with 7299 images systematically categorized into 6 classes, collected under diverse lighting conditions, camera angles, and capture settings. Using TrueSkin, we benchmark existing recognition and generation approaches, revealing substantial biases: LMMs tend to misclassify intermediate skin tones as lighter ones, whereas generative models struggle to accurately produce specified skin tones when influenced by inherent biases from unrelated attributes in the prompts, such as hairstyle or environmental context. We further demonstrate that training a recognition model on TrueSkin improves classification accuracy by more than 20\% compared to LMMs and conventional approaches, and fine-tuning with TrueSkin significantly improves skin tone fidelity in image generation models. Our findings highlight the need for comprehensive datasets like TrueSkin, which not only serves as a benchmark for evaluating existing models but also provides a valuable training resource to enhance fairness and accuracy in skin tone recognition and generation tasks.
Authors: BaiChen Fan, Yuanxi Cui, Jian Li, Qin Wang, Shibo Zhao, Muqing Cao, Sifan Zhou
Abstract: LiDAR-based 3D single object tracking (3D SOT) is a critical task in robotics and autonomous systems. Existing methods typically follow frame-wise motion estimation or a sequence-based paradigm. However, the two-frame methods are efficient but lack long-term temporal context, making them vulnerable in sparse or occluded scenes, while sequence-based methods that process multiple point clouds gain robustness at a significant computational cost. To resolve this dilemma, we propose a novel trajectory-based paradigm and its instantiation, TrajTrack. TrajTrack is a lightweight framework that enhances a base two-frame tracker by implicitly learning motion continuity from historical bounding box trajectories alone-without requiring additional, costly point cloud inputs. It first generates a fast, explicit motion proposal and then uses an implicit motion modeling module to predict the future trajectory, which in turn refines and corrects the initial proposal. Extensive experiments on the large-scale NuScenes benchmark show that TrajTrack achieves new state-of-the-art performance, dramatically improving tracking precision by 3.02% over a strong baseline while running at 55 FPS. Besides, we also demonstrate the strong generalizability of TrajTrack across different base trackers. Code is available at https://github.com/FiBonaCci225/TrajTrack.
Authors: Diogo Mendon\c{c}a, Tiago Barros, Cristiano Premebida, Urbano J. Nunes
Abstract: Autonomous-driving perception systems require robust Multi-Object Tracking (MOT) to operate reliably in dynamic environments. MOT maintains consistent object identities across frames while preserving spatial accuracy. Recent foundation models, such as SAM2, provide promptable video segmentation without task-specific fine-tuning. However, their direct application to Multi-Object Tracking and Segmentation (MOTS) remains limited by the absence of explicit identity management mechanisms and by growing memory requirements during tracking. This work introduces Seg2Track-SAM2, a framework that integrates pretrained object detectors with SAM2 and a dedicated Seg2Track module to support track initialization, data association, and track refinement. The method operates without dataset-specific fine-tuning and remains detector-agnostic. Experimental evaluation on the KITTI MOTS and MOTS Challenge benchmarks shows that Seg2Track-SAM2 ranks fourth overall in both datasets while achieving the highest association accuracy (AssA) among compared methods. In addition, a sliding-window memory strategy reduces memory usage by up to 75% with minimal impact on tracking performance, enabling deployment under resource constraints. Together, these results indicate that Seg2Track-SAM2 improves identity consistency and memory efficiency in MOTS without requiring dataset-specific training. The code is available at https://github.com/hcmr-lab/Seg2Track-SAM2.
Authors: Hanshuai Cui, Zhiqing Tang, Zhifei Xu, Zhi Yao, Wenyi Zeng, Weijia Jia
Abstract: Recent advancements in Diffusion Transformers (DiTs) have established them as the state-of-the-art method for video generation. However, their inherently sequential denoising process results in inevitable latency, limiting real-world applicability. Existing acceleration methods either compromise visual quality due to architectural modifications or fail to reuse intermediate features at proper granularity. Our analysis reveals that DiT blocks are the primary contributors to inference latency. Across diffusion timesteps, the feature variations of DiT blocks exhibit a U-shaped pattern with high similarity during intermediate timesteps, which suggests substantial computational redundancy. In this paper, we propose Block-Wise Caching (BWCache), a training-free method to accelerate DiT-based video generation. BWCache dynamically caches and reuses features from DiT blocks across diffusion timesteps. Furthermore, we introduce a similarity indicator that triggers feature reuse only when the differences between block features at adjacent timesteps fall below a threshold, thereby minimizing redundant computations while maintaining visual fidelity. Extensive experiments on several video diffusion models demonstrate that BWCache achieves up to 6$\times$ speedup with comparable visual quality.
Authors: Junhao Jia, Yunyou Liu, Cheng Yang, Yifei Sun, Feiwei Qin, Changmiao Wang, Yong Peng
Abstract: Functional magnetic resonance imaging (fMRI) reveals complex brain functional networks with hierarchical topologies crucial for cognitive processing. Standard Euclidean Graph Neural Networks (GNNs) often struggle to represent these hierarchical structures without high distortion due to inherent spatial constraints. We propose Brain-HGCN, a geometric deep learning framework based on hyperbolic geometry, which leverages negatively curved space to model brain network hierarchy with high fidelity. Grounded in the Lorentz model, our framework employs a novel hyperbolic graph attention layer with a signed aggregation mechanism to distinctly process excitatory and inhibitory connections. Furthermore, we learn robust graph-level representations via a geometrically principled Fr\'echet mean for graph readout. Experiments on two large-scale fMRI datasets for psychiatric disorder classification demonstrate that Brain-HGCN significantly outperforms state-of-the-art Euclidean baselines. This work highlights the potential of hyperbolic GNNs in computational psychiatry by pioneering a new geometric paradigm for fMRI analysis.
Authors: Junhao Jia, Yunyou Liu, Yifei Sun, Huangwei Chen, Feiwei Qin, Changmiao Wang, Yong Peng
Abstract: Nonlinear manifolds are pervasive in deep visual features, where Euclidean distances can misrepresent true similarity. This mismatch is particularly detrimental to prototype-based interpretable fine-grained recognition, where even subtle semantic distinctions are crucial. To mitigate this issue, this work presents a novel paradigm for prototype-based recognition by grounding similarity in the intrinsic geometry of deep features. Concretely, we distill the latent manifold structure of each class into a diffusion space and, critically, devise a differentiable Nystr\"{o}m interpolation to make this geometry accessible to both unseen samples and learnable prototypes. To maintain efficiency, we employ compact per-class landmark sets with periodic updates. This strategy keeps the embedding synchronized with the evolving backbone, enabling fast inference at scale. Comprehensive experiments on two benchmark datasets demonstrate that our GeoProto yields prototypes focusing on semantically corresponding parts, significantly outperforming Euclidean prototype networks.
Authors: Shilin Lu, Zhuming Lian, Zihan Zhou, Shaocong Zhang, Chen Zhao, Adams Wai-Kin Kong
Abstract: Image composition aims to seamlessly insert a user-specified object into a new scene, but existing models struggle with complex lighting (e.g., accurate shadows, water reflections) and diverse, high-resolution inputs. Modern text-to-image diffusion models (e.g., SD3.5, FLUX) already encode essential physical and resolution priors, yet lack a framework to unleash them without resorting to latent inversion, which often locks object poses into contextually inappropriate orientations, or brittle attention surgery. We propose SHINE, a training-free framework for Seamless, High-fidelity Insertion with Neutralized Errors. SHINE introduces manifold-steered anchor loss, leveraging pretrained customization adapters (e.g., IP-Adapter) to guide latents for faithful subject representation while preserving background integrity. Degradation-suppression guidance and adaptive background blending are proposed to further eliminate low-quality outputs and visible seams. To address the lack of rigorous benchmarks, we introduce ComplexCompo, featuring diverse resolutions and challenging conditions such as low lighting, strong illumination, intricate shadows, and reflective surfaces. Experiments on ComplexCompo and DreamEditBench show state-of-the-art performance on standard metrics (e.g., DINOv2) and human-aligned scores (e.g., DreamSim, ImageReward, VisionReward). Code is available at https://github.com/ZhumingLian/SHINE.
Authors: Weilun Feng, Haotong Qin, Mingqiang Wu, Chuanguang Yang, Yuqi Li, Xiangqi Li, Zhulin An, Libo Huang, Yulun Zhang, Michele Magno, Yongjun Xu
Abstract: Learning-based 3D reconstruction models, represented by Visual Geometry Grounded Transformers (VGGTs), have made remarkable progress with the use of large-scale transformers. Their prohibitive computational and memory costs severely hinder real-world deployment. Post-Training Quantization (PTQ) has become a common practice for compressing and accelerating models. However, we empirically observe that PTQ faces unique obstacles when compressing billion-scale VGGTs: the data-independent special tokens induce heavy-tailed activation distributions, while the multi-view nature of 3D data makes calibration sample selection highly unstable. This paper proposes the first Quantization framework for VGGTs, namely QuantVGGT. This mainly relies on two technical contributions: First, we introduce Dual-Smoothed Fine-Grained Quantization, which integrates pre-global Hadamard rotation and post-local channel smoothing to mitigate heavy-tailed distributions and inter-channel variance robustly. Second, we design Noise-Filtered Diverse Sampling, which filters outliers via deep-layer statistics and constructs frame-aware diverse calibration clusters to ensure stable quantization ranges. Comprehensive experiments demonstrate that QuantVGGT achieves the state-of-the-art results across different benchmarks and bit-width, surpassing the previous state-of-the-art generic quantization method with a great margin. We highlight that our 4-bit QuantVGGT can deliver a 3.7$\times$ memory reduction and 2.5$\times$ acceleration in real-hardware inference, while maintaining reconstruction accuracy above 98\% of its full-precision counterpart. This demonstrates the vast advantages and practicality of QuantVGGT in resource-constrained scenarios. Our code is released in https://github.com/wlfeng0509/QuantVGGT.
Authors: Jian Liu, Chunshi Wang, Song Guo, Haohan Weng, Zhen Zhou, Zhiqi Li, Jiaao Yu, Yiling Zhu, Jing Xu, Biwen Lei, Zhuo Chen, Chunchao Guo
Abstract: The generation of quadrilateral-dominant meshes is a cornerstone of professional 3D content creation. However, existing generative models generate quad meshes by first generating triangle meshes and then merging triangles into quadrilaterals with some specific rules, which typically produces quad meshes with poor topology. In this paper, we introduce QuadGPT, the first autoregressive framework for generating quadrilateral meshes in an end-to-end manner. QuadGPT formulates this as a sequence prediction paradigm, distinguished by two key innovations: a unified tokenization method to handle mixed topologies of triangles and quadrilaterals, and a specialized Reinforcement Learning fine-tuning method tDPO for better generation quality. Extensive experiments demonstrate that QuadGPT significantly surpasses previous triangle-to-quad conversion pipelines in both geometric accuracy and topological quality. Our work establishes a new benchmark for native quad-mesh generation and showcases the power of combining large-scale autoregressive models with topology-aware RL refinement for creating structured 3D assets.
Authors: Wenyi Gong, Mieszko Lis
Abstract: Many modern ViT backbones adopt spatial architectural designs, such as window attention, decomposed relative positional embeddings in SAM, and RoPE in DINOv3. Such architectures impose new challenges on token reduction, as the vast majority of existing methods fail to preserve the spatial structure these architectures depend on. In this paper, we introduce a simple yet effective token merging method that maintains spatial integrity, enabling seamless compatibility with spatial architectures. We reconcile two seemingly conflicting requirements: (i)exploiting the uneven information distribution across the spatial layout while (ii)preserving the spatial structure post-merging. Our approach employs (i)a 2D reduction strategy to enforce structured token layouts, (ii)a spatial-aware merging algorithm that maintains relative token positions, and (iii)a novel max-magnitude-per-dimension token representation that preserves salient features. Our method demonstrates strong performance both off-the-shelf and with fine-tuning, achieving state-of-the-art results on spatial and non-spatial architectures across various vision tasks. Specifically, we achieve 1.25x speedup on SAM-H with only 0.7% mIOU drop evaluated on COCO off-the-shelf, and 1.15x speedup on DeiT-B with no top-1 accuracy drop on ImageNet within just one epoch of fine-tuning.
Authors: Daiqing Wu, Dongbao Yang, Sicheng Zhao, Can Ma, Yu Zhou
Abstract: Recently, Multimodal Large Language Models (MLLMs) have achieved exceptional performance across diverse tasks, continually surpassing previous expectations regarding their capabilities. Nevertheless, their proficiency in perceiving emotions from images remains debated, with studies yielding divergent results in zero-shot scenarios. We argue that this inconsistency stems partly from constraints in existing evaluation methods, including the oversight of plausible responses, limited emotional taxonomies, neglect of contextual factors, and labor-intensive annotations. To facilitate customized visual emotion evaluation for MLLMs, we propose an Emotion Statement Judgment task that overcomes these constraints. Complementing this task, we devise an automated pipeline that efficiently constructs emotion-centric statements with minimal human effort. Through systematically evaluating prevailing MLLMs, our study showcases their stronger performance in emotion interpretation and context-based emotion judgment, while revealing relative limitations in comprehending perception subjectivity. When compared to humans, even top-performing MLLMs like GPT4o demonstrate remarkable performance gaps, underscoring key areas for future improvement. By developing a fundamental evaluation framework and conducting a comprehensive MLLM assessment, we hope this work contributes to advancing emotional intelligence in MLLMs. Project page: https://github.com/wdqqdw/MVEI.
Authors: Arman Akbari, Jian Gao, Yifei Zou, Mei Yang, Jinru Duan, Dmitrii Torbunov, Yanzhi Wang, Yihui Ren, Xuan Zhang
Abstract: Engineering design operates through hierarchical abstraction from system specifications to component implementations, requiring visual understanding coupled with mathematical reasoning at each level. While Multi-modal Large Language Models (MLLMs) excel at natural image tasks, their ability to extract mathematical models from technical diagrams remains unexplored. We present \textbf{CircuitSense}, a comprehensive benchmark evaluating circuit understanding across this hierarchy through 8,006+ problems spanning component-level schematics to system-level block diagrams. Our benchmark uniquely examines the complete engineering workflow: Perception, Analysis, and Design, with a particular emphasis on the critical but underexplored capability of deriving symbolic equations from visual inputs. We introduce a hierarchical synthetic generation pipeline consisting of a grid-based schematic generator and a block diagram generator with auto-derived symbolic equation labels. Comprehensive evaluation of six state-of-the-art MLLMs, including both closed-source and open-source models, reveals fundamental limitations in visual-to-mathematical reasoning. Closed-source models achieve over 85\% accuracy on perception tasks involving component recognition and topology identification, yet their performance on symbolic derivation and analytical reasoning falls below 19\%, exposing a critical gap between visual parsing and symbolic reasoning. Models with stronger symbolic reasoning capabilities consistently achieve higher design task accuracy, confirming the fundamental role of mathematical understanding in circuit synthesis and establishing symbolic reasoning as the key metric for engineering competence.
Authors: Pinyuan Feng, Hossein Adeli, Wenxuan Guo, Fan Cheng, Ethan Hwang, Nikolaus Kriegeskorte
Abstract: Recent work has demonstrated that complex visual stimuli can be decoded from human brain activity using deep generative models, offering new ways to probe how the brain represents real-world scenes. However, many existing approaches first map brain signals into intermediate image or text feature spaces before guiding the generative process, which obscures the contributions of different brain areas to the final reconstruction output. In this work, we propose NeuroAdapter, a visual decoding framework that directly conditions a latent diffusion model on brain representations, bypassing the need for intermediate feature spaces. Our method demonstrates competitive visual reconstruction quality on public fMRI datasets compared to prior work, while providing greater transparency into how brain signals drive visual reconstruction. To this end, we introduce an Image-Brain BI-directional interpretability framework (IBBI) that analyzes cross-attention patterns across diffusion denoising steps to reveal how different cortical areas influence the unfolding generative trajectory. Our work highlights the potential of end-to-end brain-to-image reconstruction and establishes a path for interpretable neural decoding.
Authors: Wei Pan, Huiguo He, Hiuyi Cheng, Yilin Shi, Lianwen Jin
Abstract: Deep generative models have advanced text-to-online handwriting generation (TOHG), which aims to synthesize realistic pen trajectories conditioned on textual input and style references. However, most existing methods still primarily focus on character- or word-level generation, resulting in inefficiency and a lack of holistic structural modeling when applied to full text lines. To address these issues, we propose DiffInk, the first latent diffusion Transformer framework for full-line handwriting generation. We first introduce InkVAE, a novel sequential variational autoencoder enhanced with two complementary latent-space regularization losses: (1) an OCR-based loss enforcing glyph-level accuracy, and (2) a style-classification loss preserving writing style. This dual regularization yields a semantically structured latent space where character content and writer styles are effectively disentangled. We then introduce InkDiT, a novel latent diffusion Transformer that integrates target text and reference styles to generate coherent pen trajectories. Experimental results demonstrate that DiffInk outperforms existing state-of-the-art (SOTA) methods in both glyph accuracy and style fidelity, while significantly improving generation efficiency.
Authors: Weilun Feng, Chuanguang Yang, Haotong Qin, Mingqiang Wu, Yuqi Li, Xiangqi Li, Zhulin An, Libo Huang, Yulun Zhang, Michele Magno, Yongjun Xu
Abstract: Diffusion transformers exhibit remarkable video generation capability, yet their prohibitive computational and memory costs hinder practical deployment. Model quantization and attention sparsification are two promising directions for compression, but each alone suffers severe performance degradation under aggressive compression. Combining them promises compounded efficiency gains, but naive integration is ineffective. The sparsity-induced information loss exacerbates quantization noise, leading to amplified attention shifts. To address this, we propose \textbf{QuantSparse}, a unified framework that integrates model quantization with attention sparsification. Specifically, we introduce \textit{Multi-Scale Salient Attention Distillation}, which leverages both global structural guidance and local salient supervision to mitigate quantization-induced bias. In addition, we develop \textit{Second-Order Sparse Attention Reparameterization}, which exploits the temporal stability of second-order residuals to efficiently recover information lost under sparsity. Experiments on HunyuanVideo-13B demonstrate that QuantSparse achieves 20.88 PSNR, substantially outperforming the state-of-the-art quantization baseline Q-VDiT (16.85 PSNR), while simultaneously delivering a \textbf{3.68$\times$} reduction in storage and \textbf{1.88$\times$} acceleration in end-to-end inference. Our code will be released in https://github.com/wlfeng0509/QuantSparse.
Authors: Muleilan Pei, Shaoshuai Shi, Shaojie Shen
Abstract: Scalable and realistic simulation of multi-agent traffic behavior is critical for advancing autonomous driving technologies. Although existing data-driven simulators have made significant strides in this domain, they predominantly rely on supervised learning to align simulated distributions with real-world driving scenarios. A persistent challenge, however, lies in the distributional shift that arises between training and testing, which often undermines model generalization in unseen environments. To address this limitation, we propose SMART-R1, a novel R1-style reinforcement fine-tuning paradigm tailored for next-token prediction models to better align agent behavior with human preferences and evaluation metrics. Our approach introduces a metric-oriented policy optimization algorithm to improve distribution alignment and an iterative "SFT-RFT-SFT" training strategy that alternates between Supervised Fine-Tuning (SFT) and Reinforcement Fine-Tuning (RFT) to maximize performance gains. Extensive experiments on the large-scale Waymo Open Motion Dataset (WOMD) validate the effectiveness of this simple yet powerful R1-style training framework in enhancing foundation models. The results on the Waymo Open Sim Agents Challenge (WOSAC) showcase that SMART-R1 achieves state-of-the-art performance with an overall realism meta score of 0.7858, ranking first on the leaderboard at the time of submission.
Authors: Jitai Hao, Hao Liu, Xinyan Xiao, Qiang Huang, Jun Yu
Abstract: Unified Multimodal Models (UMMs) built on shared autoregressive (AR) transformers are attractive for their architectural simplicity. However, we identify a critical limitation: when trained on multimodal inputs, modality-shared transformers suffer from severe gradient conflicts between vision and text, particularly in shallow and deep layers. We trace this issue to the fundamentally different low-level statistical properties of images and text, while noting that conflicts diminish in middle layers where representations become more abstract and semantically aligned. To overcome this challenge, we propose Uni-X, a two-end-separated, middle-shared architecture. Uni-X dedicates its initial and final layers to modality-specific processing, while maintaining shared parameters in the middle layers for high-level semantic fusion. This X-shaped design not only eliminates gradient conflicts at both ends but also further alleviates residual conflicts in the shared layers. Extensive experiments validate the effectiveness of Uni-X. Under identical training conditions, Uni-X achieves superior training efficiency compared to strong baselines. When scaled to 3B parameters with larger training data, Uni-X matches or surpasses 7B AR-based UMMs, achieving a GenEval score of 82 for image generation alongside strong performance in text and vision understanding tasks. These results establish Uni-X as a parameter-efficient and scalable foundation for future unified multimodal modeling. Our code is available at https://github.com/CURRENTF/Uni-X
Authors: Haijier Chen, Bo Xu, Shoujian Zhang, Haoze Liu, Jiaxuan Lin, Jingrong Wang
Abstract: Recent developments in Multimodal Large Language Models (MLLMs) have significantly improved Vision-Language (VL) reasoning in 2D domains. However, extending these capabilities to 3D scene understanding remains a major challenge. Existing 3D Multimodal Large Language Models (3D-MLLMs) often depend on 3D data inputs, which limits scalability and generalization. To address this limitation, we propose Vid-LLM, a video-based 3D-MLLM that directly processes video inputs without requiring external 3D data, making it practical for real-world deployment. In our method, the geometric prior are directly used to improve the performance of the sceen perception. To integrate the geometric cues into the MLLM compactly, we design a Cross-Task Adapter (CTA) module to align the 3D geometric priors with the vision-language representations. To ensure geometric consistency and integrity, we introduce a Metric Depth Model that recovers real-scale geometry from the reconstruction outputs. Finally, the model is fine-tuned with a two-stage distillation optimization strategy, realizing fast convergence and stabilizes training. Extensive experiments across diverse benchmarks verified the effectiveness of our method on 3D Question Answering, 3D Dense Captioning and 3D Visual Grounding tasks, demonstrating the superior multi-task capabilities.
Authors: Yuyou Zhang, Radu Corcodel, Chiori Hori, Anoop Cherian, Ding Zhao
Abstract: We present SpinBench, a cognitively grounded diagnostic benchmark for evaluating spatial reasoning in vision language models (VLMs). SpinBench is designed around the core challenge of spatial reasoning: perspective taking, the ability to reason about how scenes and object relations change under viewpoint transformation. Since perspective taking requires multiple cognitive capabilities, such as recognizing objects across views, relative positions grounding, and mentally simulating transformations, SpinBench introduces a set of fine-grained diagnostic categories. Our categories target translation, rotation, object relative pose, and viewpoint change, and are progressively structured so that single-object simpler tasks scaffold toward the most demanding multi-object perspective-taking setting. We evaluate 43 state-of-the-art VLMs, both proprietary and open source. Results reveal systematic weaknesses: strong egocentric bias, poor rotational understanding, and inconsistencies under symmetrical and syntactic reformulations. Scaling analysis shows both smooth improvements and emergent capabilities. While human subjects achieve high accuracy (91.2\%), task difficulty as measured by human response time shows strong correlation with VLM accuracy, indicating that SpinBench captures spatial reasoning challenges shared across humans and VLMs. We believe SpinBench provides critical insights into spatial reasoning in VLMs and highlights key gaps in their ability to reason about physical space. Our website can be found at https://spinbench25.github.io/.
Authors: Yuan Zhao, Youwei Pang, Lihe Zhang, Hanqi Liu, Jiaming Zuo, Huchuan Lu, Xiaoqi Zhao
Abstract: Existing anomaly detection (AD) methods often treat the modality and class as independent factors. Although this paradigm has enriched the development of AD research branches and produced many specialized models, it has also led to fragmented solutions and excessive memory overhead. Moreover, reconstruction-based multi-class approaches typically rely on shared decoding paths, which struggle to handle large variations across domains, resulting in distorted normality boundaries, domain interference, and high false alarm rates. To address these limitations, we propose UniMMAD, a unified framework for multi-modal and multi-class anomaly detection. At the core of UniMMAD is a Mixture-of-Experts (MoE)-driven feature decompression mechanism, which enables adaptive and disentangled reconstruction tailored to specific domains. This process is guided by a ``general to specific'' paradigm. In the encoding stage, multi-modal inputs of varying combinations are compressed into compact, general-purpose features. The encoder incorporates a feature compression module to suppress latent anomalies, encourage cross-modal interaction, and avoid shortcut learning. In the decoding stage, the general features are decompressed into modality-specific and class-specific forms via a sparsely-gated cross MoE, which dynamically selects expert pathways based on input modality and class. To further improve efficiency, we design a grouped dynamic filtering mechanism and a MoE-in-MoE structure, reducing parameter usage by 75\% while maintaining sparse activation and fast inference. UniMMAD achieves state-of-the-art performance on 9 anomaly detection datasets, spanning 3 fields, 12 modalities, and 66 classes. The source code will be available at https://github.com/yuanzhao-CVLAB/UniMMAD.
Authors: Edoardo Bianchi, Jacopo Staiano, Antonio Liotta
Abstract: Most existing approaches formulate action quality assessment and skill proficiency estimation as discriminative prediction tasks, typically producing discrete labels or scores without explicitly modeling the reasoning process underlying the assessment. We instead reformulate the problem as generative vision-language modeling, introducing ProfVLM, a parameter-efficient vision-language model that jointly predicts proficiency levels and generates expert-like natural language feedback from multi-view videos. ProfVLM leverages conditional language generation to provide actionable insights along with quantitative evaluation scores. Central to our method is an AttentiveGatedProjector that dynamically fuses and projects multi-view egocentric and exocentric features from a frozen TimeSformer backbone into a language model fine-tuned for feedback generation. Trained on EgoExo4D with expert commentaries, ProfVLM surpasses state-of-the-art methods while using up to 20x fewer parameters and reducing training time by up to 60% compared to existing classification-based methods. By providing natural language critiques aligned with performance levels, this work shows that generative vision-language modeling offers a powerful and efficient paradigm shift for interpretable action quality assessment.
Authors: Keming Wu, Sicong Jiang, Max Ku, Ping Nie, Minghao Liu, Wenhu Chen
Abstract: Recently, we have witnessed great progress in image editing with natural language instructions. Several closed-source models like GPT-Image-1, Seedream, and Google-Nano-Banana have shown highly promising progress. However, the open-source models are still lagging. The main bottleneck is the lack of a reliable reward model to scale up high-quality synthetic training data. To address this critical bottleneck, we built EditReward, trained with our new large-scale human preference dataset, meticulously annotated by trained experts following a rigorous protocol containing over 200K preference pairs. EditReward demonstrates superior alignment with human preferences in instruction-guided image editing tasks. Experiments show that EditReward achieves state-of-the-art human correlation on established benchmarks such as GenAI-Bench, AURORA-Bench, ImagenHub, and our new EditReward-Bench, outperforming a wide range of VLM-as-judge models. Furthermore, we use EditReward to select a high-quality subset from the existing noisy ShareGPT-4o-Image dataset. We train Step1X-Edit on the selected subset, which shows significant improvement over training on the full set. This demonstrates EditReward's ability to serve as a reward model to scale up high-quality training data for image editing. Furthermore, its strong alignment suggests potential for advanced applications like reinforcement learning-based post-training and test-time scaling of image editing models. EditReward with its training dataset will be released to help the community build more high-quality image editing training datasets.
Authors: Hanzhou Liu, Jia Huang, Mi Lu, Srikanth Saripalli, Peng Jiang
Abstract: We present Stylos, a single-forward 3D Gaussian framework for 3D style transfer that operates on unposed content, from a single image to a multi-view collection, conditioned on a separate reference style image. Stylos synthesizes a stylized 3D Gaussian scene without per-scene optimization or precomputed poses, achieving geometry-aware, view-consistent stylization that generalizes to unseen categories, scenes, and styles. At its core, Stylos adopts a Transformer backbone with two pathways: geometry predictions retain self-attention to preserve geometric fidelity, while style is injected via global cross-attention to enforce visual consistency across views. With the addition of a voxel-based 3D style loss that aligns aggregated scene features to style statistics, Stylos enforces view-consistent stylization while preserving geometry. Experiments across multiple datasets demonstrate that Stylos delivers high-quality zero-shot stylization, highlighting the effectiveness of global style-content coupling, the proposed 3D style loss, and the scalability of our framework from single view to large-scale multi-view settings. Our codes are available at https://github.com/HanzhouLiu/Stylos.
Authors: Yuchen Song, Andong Chen, Wenxin Zhu, Kehai Chen, Xuefeng Bai, Muyun Yang, Tiejun Zhao
Abstract: Cultural awareness capabilities have emerged as a critical capability for Multimodal Large Language Models (MLLMs). However, current benchmarks lack progressed difficulty in their task design and are deficient in cross-lingual tasks. Moreover, current benchmarks often use real-world images. Each real-world image typically contains one culture, making these benchmarks relatively easy for MLLMs. Based on this, we propose C$^3$B (Comics Cross-Cultural Benchmark), a novel multicultural, multitask and multilingual cultural awareness capabilities benchmark. C$^3$B comprises over 2000 images and over 18000 QA pairs, constructed on three tasks with progressed difficulties, from basic visual recognition to higher-level cultural conflict understanding, and finally to cultural content generation. We conducted evaluations on 11 open-source MLLMs, revealing a significant performance gap between MLLMs and human performance. The gap demonstrates that C$^3$B poses substantial challenges for current MLLMs, encouraging future research to advance the cultural awareness capabilities of MLLMs.
Authors: Guozhen Zhang, Haiguang Wang, Chunyu Wang, Yuan Zhou, Qinglin Lu, Limin Wang
Abstract: Video frame interpolation (VFI), which generates intermediate frames from given start and end frames, has become a fundamental function in video generation applications. However, existing generative VFI methods are constrained to synthesize a fixed number of intermediate frames, lacking the flexibility to adjust generated frame rates or total sequence duration. In this work, we present ArbInterp, a novel generative VFI framework that enables efficient interpolation at any timestamp and of any length. Specifically, to support interpolation at any timestamp, we propose the Timestamp-aware Rotary Position Embedding (TaRoPE), which modulates positions in temporal RoPE to align generated frames with target normalized timestamps. This design enables fine-grained control over frame timestamps, addressing the inflexibility of fixed-position paradigms in prior work. For any-length interpolation, we decompose long-sequence generation into segment-wise frame synthesis. We further design a novel appearance-motion decoupled conditioning strategy: it leverages prior segment endpoints to enforce appearance consistency and temporal semantics to maintain motion coherence, ensuring seamless spatiotemporal transitions across segments. Experimentally, we build comprehensive benchmarks for multi-scale frame interpolation (2x to 32x) to assess generalizability across arbitrary interpolation factors. Results show that ArbInterp outperforms prior methods across all scenarios with higher fidelity and more seamless spatiotemporal continuity. Project website: https://mcg-nju.github.io/ArbInterp-Web/.
Authors: Alessio Spagnoletti, Andr\'es Almansa, Marcelo Pereyra
Abstract: Computational imaging methods increasingly rely on powerful generative diffusion models to tackle challenging image restoration tasks. In particular, state-of-the-art zero-shot image inverse solvers leverage distilled text-to-image latent diffusion models (LDMs) to achieve unprecedented accuracy and perceptual quality with high computational efficiency. However, extending these advances to high-definition video restoration remains a significant challenge, due to the need to recover fine spatial detail while capturing subtle temporal dependencies. Consequently, methods that naively apply image-based LDM priors on a frame-by-frame basis often result in temporally inconsistent reconstructions. We address this challenge by leveraging recent advances in Video Consistency Models (VCMs), which distill video latent diffusion models into fast generators that explicitly capture temporal causality. Building on this foundation, we propose LVTINO, the first zero-shot or plug-and-play inverse solver for high definition video restoration with priors encoded by VCMs. Our conditioning mechanism bypasses the need for automatic differentiation and achieves state-of-the-art video reconstruction quality with only a few neural function evaluations, while ensuring strong measurement consistency and smooth temporal transitions across frames. Extensive experiments on a diverse set of video inverse problems show significant perceptual improvements over current state-of-the-art methods that apply image LDMs frame by frame, establishing a new benchmark in both reconstruction fidelity and computational efficiency. The code is available on GitHub.
Authors: R\u{a}zvan-Andrei Mati\c{s}an, Vincent Tao Hu, Grigory Bartosh, Bj\"orn Ommer, Cees G. M. Snoek, Max Welling, Jan-Willem van de Meent, Mohammad Mahdi Derakhshani, Floor Eijkelboom
Abstract: We introduce Purrception, a variational flow matching approach for vector-quantized image generation that provides explicit categorical supervision while maintaining continuous transport dynamics. Our method adapts Variational Flow Matching to vector-quantized latents by learning categorical posteriors over codebook indices while computing velocity fields in the continuous embedding space. This combines the geometric awareness of continuous methods with the discrete supervision of categorical approaches, enabling uncertainty quantification over plausible codes and temperature-controlled generation. We evaluate Purrception on ImageNet-1k 256x256 generation. Training converges faster than both continuous flow matching and discrete flow matching baselines while achieving competitive FID scores with state-of-the-art models. This demonstrates that Variational Flow Matching can effectively bridge continuous transport and discrete supervision for improved training efficiency in image generation.
Authors: Zihan Zhou, Shilin Lu, Shuli Leng, Shaocong Zhang, Zhuming Lian, Xinlei Yu, Adams Wai-Kin Kong
Abstract: Drag-based image editing has long suffered from distortions in the target region, largely because the priors of earlier base models, Stable Diffusion, are insufficient to project optimized latents back onto the natural image manifold. With the shift from UNet-based DDPMs to more scalable DiT with flow matching (e.g., SD3.5, FLUX), generative priors have become significantly stronger, enabling advances across diverse editing tasks. However, drag-based editing has yet to benefit from these stronger priors. This work proposes the first framework to effectively harness FLUX's rich prior for drag-based editing, dubbed DragFlow, achieving substantial gains over baselines. We first show that directly applying point-based drag editing to DiTs performs poorly: unlike the highly compressed features of UNets, DiT features are insufficiently structured to provide reliable guidance for point-wise motion supervision. To overcome this limitation, DragFlow introduces a region-based editing paradigm, where affine transformations enable richer and more consistent feature supervision. Additionally, we integrate pretrained open-domain personalization adapters (e.g., IP-Adapter) to enhance subject consistency, while preserving background fidelity through gradient mask-based hard constraints. Multimodal large language models (MLLMs) are further employed to resolve task ambiguities. For evaluation, we curate a novel Region-based Dragging benchmark (ReD Bench) featuring region-level dragging instructions. Extensive experiments on DragBench-DR and ReD Bench show that DragFlow surpasses both point-based and region-based baselines, setting a new state-of-the-art in drag-based image editing. Code and dataset are available at https://github.com/Edennnnnnnnnn/DragFlow.
Authors: Deheng Zhang, Yuqian Fu, Runyi Yang, Yang Miao, Tianwen Qian, Xu Zheng, Guolei Sun, Ajad Chhatkuli, Xuanjing Huang, Yu-Gang Jiang, Luc Van Gool, Danda Pani Paudel
Abstract: Most existing benchmarks for understanding egocentric vision focus primarily on daytime scenarios, overlooking the low-light conditions that are inevitable in real-world applications. To investigate this gap, we present EgoNight, the first comprehensive benchmark for nighttime egocentric vision, with visual question answering (VQA) as the core task. A key feature of EgoNight is the introduction of day-night aligned videos, which enhance night annotation quality using the daytime data and reveal clear performance gaps between lighting conditions. To achieve this, we collect both synthetic videos rendered by Blender and real-world recordings, ensuring that scenes and actions are visually and temporally aligned. Leveraging these paired videos, we construct EgoNight-VQA, supported by a novel day-augmented night auto-labeling engine and refinement through extensive human verification. Each QA pair is double-checked by annotators for reliability. In total, EgoNight-VQA contains 3658 QA pairs across 90 videos, spanning 12 diverse QA types, with more than 300 hours of human work. Evaluations of state-of-the-art multimodal large language models (MLLMs) reveal substantial performance drops when transferring from day to night, underscoring the challenges of reasoning under low-light conditions. Beyond VQA, EgoNight also introduces two auxiliary tasks, day-night correspondence retrieval and egocentric depth estimation at night, that further explore the boundaries of existing models. We believe EgoNight-VQA provides a strong foundation for advancing application-driven egocentric vision research and for developing models that generalize across illumination domains. The code and data can be found at https://github.com/dehezhang2/EgoNight.
Authors: Yike Wu, Yiwei Wang, Yujun Cai
Abstract: While Large Vision-Language Models (LVLMs) achieve strong performance in multimodal tasks, hallucinations continue to hinder their reliability. Among the three categories of hallucinations, which include object, attribute, and relation, relation hallucinations account for the largest proportion but have received the least attention. To address this issue, we propose ChainMPQ (Multi-Perspective Questions guided Interleaved Chain of Image and Text), a training-free method that improves relational inference in LVLMs by utilizing accumulated textual and visual memories. ChainMPQ first extracts subject and object keywords from the question to enhance the corresponding image regions. It then constructs multi-perspective questions that focus on the three core components of a relationship: the subject, the object, and the relation that links them. These questions are sequentially input to the model, with textual and visual memories from earlier steps providing supporting context for subsequent ones, thereby forming an interleaved chain of images and text that guides progressive relational reasoning. Experiments on multiple LVLMs and benchmarks show that ChainMPQ substantially reduces relation hallucinations, while ablation studies further validate the effectiveness of its three core modules.
Authors: Teng Wang, Haojun Jiang, Yuxuan Wang, Zhenguo Sun, Yujiao Deng, Shiji Song, Gao Huang
Abstract: Echocardiography is a critical tool for detecting heart diseases, yet its steep operational difficulty causes a shortage of skilled personnel. Probe guidance systems, which assist in acquiring high-quality images, offer a promising solution to lower this operational barrier. However, robust probe guidance remains challenging due to significant individual variability. This variability manifests as differences in low-level features within two-dimensional (2D) images, which complicates image feature understanding, and differences in individual three-dimensional (3D) structures, which poses challenges for precise navigation. To address these challenges, we first propose leveraging the robust image representations learned by ultrasound foundation models from vast datasets. Yet, applying these models to probe navigation is non-trivial due to their lack of understanding of individual 3D structures. To this end, we meticulously design a Vision-Action Adapter (VA-Adapter) to online inject the capability of understanding individual 3D structures. Specifically, by embedding the VA-Adapter into the foundation model's image encoder, the model can infer cardiac anatomy from historical vision-action sequences, mimicking the cognitive process of a sonographer. Extensive experiments on a dataset with over 1.31M samples demonstrate that the VA-Adapter outperforms strong probe guidance models while requiring approximately 33 times fewer trained parameters.
Authors: Leigang Qu, Ziyang Wang, Na Zheng, Wenjie Wang, Liqiang Nie, Tat-Seng Chua
Abstract: Video Foundation Models (VFMs) exhibit remarkable visual generation performance, but struggle in compositional scenarios (e.g., motion, numeracy, and spatial relation). In this work, we introduce Test-Time Optimization and Memorization (TTOM), a training-free framework that aligns VFM outputs with spatiotemporal layouts during inference for better text-image alignment. Rather than direct intervention to latents or attention per-sample in existing work, we integrate and optimize new parameters guided by a general layout-attention objective. Furthermore, we formulate video generation within a streaming setting, and maintain historical optimization contexts with a parametric memory mechanism that supports flexible operations, such as insert, read, update, and delete. Notably, we found that TTOM disentangles compositional world knowledge, showing powerful transferability and generalization. Experimental results on the T2V-CompBench and Vbench benchmarks establish TTOM as an effective, practical, scalable, and efficient framework to achieve cross-modal alignment for compositional video generation on the fly.
Authors: Xiaohui Li, Shaobin Zhuang, Shuo Cao, Yang Yang, Yuandong Pu, Qi Qin, Siqi Luo, Bin Fu, Yihao Liu
Abstract: Generative models for Image Super-Resolution (SR) are increasingly powerful, yet their reliance on self-attention's quadratic complexity (O(N^2)) creates a major computational bottleneck. Linear Attention offers an O(N) solution, but its promise for photorealistic SR has remained largely untapped, historically hindered by a cascade of interrelated and previously unsolved challenges. This paper introduces LinearSR, a holistic framework that, for the first time, systematically overcomes these critical hurdles. Specifically, we resolve a fundamental, training instability that causes catastrophic model divergence using our novel "knee point"-based Early-Stopping Guided Fine-tuning (ESGF) strategy. Furthermore, we mitigate the classic perception-distortion trade-off with a dedicated SNR-based Mixture of Experts (MoE) architecture. Finally, we establish an effective and lightweight guidance paradigm, TAG, derived from our "precision-over-volume" principle. Our resulting LinearSR model simultaneously delivers state-of-the-art perceptual quality with exceptional efficiency. Its core diffusion forward pass (1-NFE) achieves SOTA-level speed, while its overall multi-step inference time remains highly competitive. This work provides the first robust methodology for applying Linear Attention in the photorealistic SR domain, establishing a foundational paradigm for future research in efficient generative super-resolution.
Authors: Daiki Yoshikawa, Takashi Matsubara
Abstract: Vision-language models have achieved remarkable success in multi-modal representation learning from large-scale pairs of visual scenes and linguistic descriptions. However, they still struggle to simultaneously express two distinct types of semantic structures: the hierarchy within a concept family (e.g., dog $\preceq$ mammal $\preceq$ animal) and the compositionality across different concept families (e.g., "a dog in a car" $\preceq$ dog, car). Recent works have addressed this challenge by employing hyperbolic space, which efficiently captures tree-like hierarchy, yet its suitability for representing compositionality remains unclear. To resolve this dilemma, we propose PHyCLIP, which employs an $\ell_1$-Product metric on a Cartesian product of Hyperbolic factors. With our design, intra-family hierarchies emerge within individual hyperbolic factors, and cross-family composition is captured by the $\ell_1$-product metric, analogous to a Boolean algebra. Experiments on zero-shot classification, retrieval, hierarchical classification, and compositional understanding tasks demonstrate that PHyCLIP outperforms existing single-space approaches and offers more interpretable structures in the embedding space.
Authors: Siyuan Huang, Xiaoye Qu, Yafu Li, Yun Luo, Zefeng He, Daizong Liu, Yu Cheng
Abstract: While Reinforcement Learning with Verifiable Rewards (RLVR) has advanced the reasoning capabilities of Large Vision-Language Models (LVLMs), most existing methods in multimodal reasoning neglect the critical role of visual perception within the RLVR optimization process. In this paper, we undertake a pioneering exploration of multimodal RLVR through the novel perspective of token perception, which measures the visual dependency of each generated token. With a granular analysis of Chain-of-Thought (CoT) processes, we uncover two key insights: first, token perception in a rollout trajectory is sparsely distributed, where only a small fraction of tokens have high visual dependency for visually-grounded reasoning; second, different trajectories exhibit significant divergence in their overall visual dependency. Based on these observations, we propose Visually-Perceptive Policy Optimization (VPPO), a novel policy gradient algorithm that explicitly leverages token perception to refine the learning signal. Specifically, VPPO achieves this through a dual mechanism: it reweights a trajectory's advantage by its overall visual dependency, and focuses policy updates exclusively on perceptually pivotal tokens. On a comprehensive suite of eight perception and reasoning benchmarks, VPPO demonstrates substantial gains over leading open-source RL-tuned models, with its effectiveness consistently validated across 7B and 32B model scales. Our findings not only establish a new token-level perceptual perspective for analyzing multimodal RLVR but also present a novel and effective optimization strategy to significantly enhance the multimodal reasoning capabilities of LVLMs.
Authors: Zhi-Fen He, Ren-Dong Xie, Bo Li, Bin Liu, Jin-Yan Hu
Abstract: Multi-label image recognition with incomplete labels is a challenging yet vital task in computer vision, which faces two fundamental challenges: learning semantic-aware features and recovering missing labels. In this paper, we propose a Co-learning framework for Semantic-aware features and Label recovery (CSL), designed to address both challenges in a unified learning paradigm. Specifically, we develop a semantic-related feature learning module that captures robust semantic-related representations by discovering semantic information and label correlations. Furthermore, a semantic-guided feature enhancement module is introduced to generate highly discriminative semantic-aware features by effectively aligning visual and semantic spaces. Finally, we present a collaborative learning framework that integrates semantic-aware feature learning with label recovery. This framework not only dynamically enhances the discriminability of semantic-aware features but also adaptively infers and recovers missing labels, thereby forming a mutually reinforcing mechanism between the two processes. Extensive experiments on three widely used public datasets (MS-COCO, VOC2007, and NUS-WIDE) demonstrate that CSL outperforms state-of-the-art methods for incomplete multi-label image recognition.
Authors: Zhengrong Yue, Haiyu Zhang, Xiangyu Zeng, Boyu Chen, Chenting Wang, Shaobin Zhuang, Lu Dong, Yi Wang, Limin Wang, Yali Wang
Abstract: Tokenizer is a crucial component for both visual understanding and generation. To advance toward the ultimate goal of universal modeling, recent research has focused on developing a unified tokenizer. However, existing tokenizers face a significant performance trade-off between understanding and generation, stemming from the inherent conflict between high-level semantic abstraction and low-level pixel reconstruction. To tackle this challenge, we propose a generic and unified tokenizer, namely UniFlow, by flexibly adapting any visual encoder with a concise reconstruction decoder. Specifically, we introduce layer-wise adaptive self-distillation applied to the well-pretrained visual encoders, which enables UniFlow to simultaneously inherit the strong semantic features for visual understanding and flexibly adapt to model fine-grained details for visual generation. Moreover, we propose a lightweight patch-wise pixel flow decoder, which efficiently achieves high-fidelity pixel reconstruction by modeling a conditional flow from the noisy state back to the patch-wise pixel domain. By leveraging the semantic features as visual conditions for the decoder, we effectively alleviate the training conflicts between understanding and generation. Furthermore, the patch-wise learning strategy simplifies the data distribution, thereby improving training efficiency. Extensive experiments across 13 challenging benchmarks spanning 7 widely studied visual understanding and generation tasks demonstrate that UniFlow achieves a win-win outcome. For instance, our 7B UniFlow-XL not only surpasses the 14B TokenFlow-XL by 6.05% on average understanding benchmarks, but also achieves a competitive results in both visual reconstruction and generation, surpassing UniTok by 0.15 in rFID and 0.09 in gFID (without guidance), respectively.
Authors: Jiachen Lei, Keli Liu, Julius Berner, Haiming Yu, Hongkai Zheng, Jiahong Wu, Xiangxiang Chu
Abstract: Pixel-space generative models are often more difficult to train and generally underperform compared to their latent-space counterparts, leaving a persistent performance and efficiency gap. In this paper, we introduce a novel two-stage training framework that closes this gap for pixel-space diffusion and consistency models. In the first stage, we pre-train encoders to capture meaningful semantics from clean images while aligning them with points along the same deterministic sampling trajectory, which evolves points from the prior to the data distribution. In the second stage, we integrate the encoder with a randomly initialized decoder and fine-tune the complete model end-to-end for both diffusion and consistency models. Our framework achieves state-of-the-art (SOTA) performance on ImageNet. Specifically, our diffusion model reaches an FID of 1.58 on ImageNet-256 and 2.35 on ImageNet-512 with 75 number of function evaluations (NFE) surpassing prior pixel-space methods and VAE-based counterparts by a large margin in both generation quality and training efficiency. In a direct comparison, our model significantly outperforms DiT while using only around 30\% of its training compute. Furthermore, our consistency model achieves an impressive FID of 8.82 on ImageNet-256, significantly outperforming its latent-space counterparts. This marks the first successful training of a consistency model directly on high-resolution images without relying on pre-trained VAEs or diffusion models. Our codes are available at: \href{https://github.com/AMAP-ML/EPG}{https://github.com/AMAP-ML/EPG}
URLs: https://github.com/AMAP-ML/EPG, https://github.com/AMAP-ML/EPG
Authors: Ziqi Jiang, Yanghao Wang, Long Chen
Abstract: Aligning features from different modalities, is one of the most fundamental challenges for cross-modal tasks. Although pre-trained vision-language models can achieve a general alignment between image and text, they often require parameter-efficient fine-tuning (PEFT) for further adjustment. Today's PEFT methods (e.g., prompt tuning, LoRA-based, or adapter-based) always selectively fine-tune a subset of parameters, which can slightly adjust either visual or textual features, and avoid overfitting. In this paper, we are the first to highlight that all existing PEFT methods perform one-step adjustment. It is insufficient for complex (or difficult) datasets, where features of different modalities are highly entangled. To this end, we propose the first model-agnostic multi-step adjustment approach by learning a cross-modal velocity field: Flow Matching Alignment (FMA). Specifically, to ensure the correspondence between categories during training, we first utilize a fixed coupling strategy. Then, we propose a noise augmentation strategy to alleviate the data scarcity issue. Finally, we design an early-stopping solver, which terminates the transformation process earlier, improving both efficiency and accuracy. Compared with one-step PEFT methods, FMA has the multi-step rectification ability to achieve more precise and robust alignment. Extensive results have demonstrated that FMA can consistently yield significant performance gains across various benchmarks and backbones, particularly on challenging datasets.
Authors: Mingxuan Liu, Honglin He, Elisa Ricci, Wayne Wu, Bolei Zhou
Abstract: Urban embodied AI agents, ranging from delivery robots to quadrupeds, are increasingly populating our cities, navigating chaotic streets to provide last-mile connectivity. Training such agents requires diverse, high-fidelity urban environments to scale, yet existing human-crafted or procedurally generated simulation scenes either lack scalability or fail to capture real-world complexity. We introduce UrbanVerse, a data-driven real-to-sim system that converts crowd-sourced city-tour videos into physics-aware, interactive simulation scenes. UrbanVerse consists of: (i) UrbanVerse-100K, a repository of 100k+ annotated urban 3D assets with semantic and physical attributes, and (ii) UrbanVerse-Gen, an automatic pipeline that extracts scene layouts from video and instantiates metric-scale 3D simulations using retrieved assets. Running in IsaacSim, UrbanVerse offers 160 high-quality constructed scenes from 24 countries, along with a curated benchmark of 10 artist-designed test scenes. Experiments show that UrbanVerse scenes preserve real-world semantics and layouts, achieving human-evaluated realism comparable to manually crafted scenes. In urban navigation, policies trained in UrbanVerse exhibit scaling power laws and strong generalization, improving success by +6.3% in simulation and +30.1% in zero-shot sim-to-real transfer comparing to prior methods, accomplishing a 300 m real-world mission with only two interventions.
Authors: Minglei Shi, Haolin Wang, Wenzhao Zheng, Ziyang Yuan, Xiaoshi Wu, Xintao Wang, Pengfei Wan, Jie Zhou, Jiwen Lu
Abstract: Recent progress in diffusion-based visual generation has largely relied on latent diffusion models with variational autoencoders (VAEs). While effective for high-fidelity synthesis, this VAE+diffusion paradigm suffers from limited training efficiency, slow inference, and poor transferability to broader vision tasks. These issues stem from a key limitation of VAE latent spaces: the lack of clear semantic separation and strong discriminative structure. Our analysis confirms that these properties are crucial not only for perception and understanding tasks, but also for the stable and efficient training of latent diffusion models. Motivated by this insight, we introduce SVG, a novel latent diffusion model without variational autoencoders, which leverages self-supervised representations for visual generation. SVG constructs a feature space with clear semantic discriminability by leveraging frozen DINO features, while a lightweight residual branch captures fine-grained details for high-fidelity reconstruction. Diffusion models are trained directly on this semantically structured latent space to facilitate more efficient learning. As a result, SVG enables accelerated diffusion training, supports few-step sampling, and improves generative quality. Experimental results further show that SVG preserves the semantic and discriminative capabilities of the underlying self-supervised representations, providing a principled pathway toward task-general, high-quality visual representations. Code and interpretations are available at https://howlin-wang.github.io/svg/.
Authors: Jinfeng Liu, Lingtong Kong, Mi Zhou, Jinwen Chen, Dan Xu
Abstract: We introduce Mono4DGS-HDR, the first system for reconstructing renderable 4D high dynamic range (HDR) scenes from unposed monocular low dynamic range (LDR) videos captured with alternating exposures. To tackle such a challenging problem, we present a unified framework with two-stage optimization approach based on Gaussian Splatting. The first stage learns a video HDR Gaussian representation in orthographic camera coordinate space, eliminating the need for camera poses and enabling robust initial HDR video reconstruction. The second stage transforms video Gaussians into world space and jointly refines the world Gaussians with camera poses. Furthermore, we propose a temporal luminance regularization strategy to enhance the temporal consistency of the HDR appearance. Since our task has not been studied before, we construct a new evaluation benchmark using publicly available datasets for HDR video reconstruction. Extensive experiments demonstrate that Mono4DGS-HDR significantly outperforms alternative solutions adapted from state-of-the-art methods in both rendering quality and speed.
Authors: Zhiyuan Feng, Zhaolu Kang, Qijie Wang, Zhiying Du, Jiongrui Yan, Shubin Shi, Chengbo Yuan, Huizhi Liang, Yu Deng, Qixiu Li, Rushuai Yang, Arctanx An, Leqi Zheng, Weijie Wang, Shawn Chen, Sicheng Xu, Yaobo Liang, Jiaolong Yang, Baining Guo
Abstract: Vision-language models (VLMs) are essential to Embodied AI, enabling robots to perceive, reason, and act in complex environments. They also serve as the foundation for the recent Vision-Language-Action (VLA) models. Yet most evaluations of VLMs focus on single-view settings, leaving their ability to integrate multi-view information underexplored. At the same time, multi-camera setups are increasingly standard in robotic platforms, as they provide complementary perspectives to mitigate occlusion and depth ambiguity. Whether VLMs can effectively leverage such multi-view inputs for robotic reasoning therefore remains an open question. To bridge this gap, we introduce MV-RoboBench, a benchmark specifically designed to evaluate the multi-view spatial reasoning capabilities of VLMs in robotic manipulation. MV-RoboBench consists of 1.7k manually curated QA items across eight subtasks, divided into two primary categories: spatial understanding and robotic execution. We evaluate a diverse set of existing VLMs, including both open-source and closed-source models, along with enhanced versions incorporating CoT-inspired techniques. The results show that state-of-the-art models remain far below human performance, underscoring the substantial challenges VLMs face in multi-view robotic perception. Additionally, our analysis uncovers two key findings: (i) spatial intelligence and robotic task execution are positively correlated in multi-view robotic scenarios; and (ii) strong performance on existing general-purpose single-view spatial understanding benchmarks does not reliably translate to success in the robotic spatial tasks assessed by our benchmark. We release MV-RoboBench as an open resource to foster progress in spatially grounded VLMs and VLAs, providing not only data but also a standardized evaluation protocol for multi-view embodied reasoning.
Authors: Ziheng Zhang, Xinyue Ma, Arpita Chowdhury, Elizabeth G. Campolongo, Matthew J. Thompson, Net Zhang, Samuel Stevens, Hilmar Lapp, Tanya Berger-Wolf, Yu Su, Wei-Lun Chao, Jianyang Gu
Abstract: This work investigates descriptive captions as an additional source of supervision for biological multimodal foundation models. Images and captions can be viewed as complementary samples from the latent morphospace of a species, each capturing certain biological traits. Incorporating captions during training encourages alignment with this shared latent structure, emphasizing potentially diagnostic characters while suppressing spurious correlations. The main challenge, however, lies in obtaining faithful, instance-specific captions at scale. This requirement has limited the utilization of natural language supervision in organismal biology compared with many other scientific domains. We complement this gap by generating synthetic captions with multimodal large language models (MLLMs), guided by Wikipedia-derived visual information and taxon-tailored format examples. These domain-specific contexts help reduce hallucination and yield accurate, instance-based descriptive captions. Using these captions, we train BioCAP (i.e., BioCLIP with Captions), a biological foundation model that captures rich semantics and achieves strong performance in species classification and text-image retrieval. These results demonstrate the value of descriptive captions beyond labels in bridging biological images with multimodal foundation models.
Authors: Orest Kupyn, Hirokatsu Kataoka, Christian Rupprecht
Abstract: Salient object detection exemplifies data-bounded tasks where expensive pixel-precise annotations force separate model training for related subtasks like DIS and HR-SOD. We present a method that dramatically improves generalization through large-scale synthetic data generation and ambiguity-aware architecture. We introduce S3OD, a dataset of over 139,000 high-resolution images created through our multi-modal diffusion pipeline that extracts labels from diffusion and DINO-v3 features. The iterative generation framework prioritizes challenging categories based on model performance. We propose a streamlined multi-mask decoder that handles the inherent ambiguity in salient object detection by predicting multiple valid interpretations. Models trained only on synthetic data achieve 20-50% error reduction in cross-dataset generalization, while fine-tuned versions reach state-of-the-art performance across DIS and HR-SOD benchmarks.
Authors: Rishit Dagli, Donglai Xiang, Vismay Modi, Charles Loop, Clement Fuji Tsang, Anka He Chen, Anita Hu, Gavriel State, David I. W. Levin, Maria Shugrina
Abstract: Physical simulation relies on spatially-varying mechanical properties, often laboriously hand-crafted. VoMP is a feed-forward method trained to predict Young's modulus ($E$), Poisson's ratio ($\nu$), and density ($\rho$) throughout the volume of 3D objects, in any representation that can be rendered and voxelized. VoMP aggregates per-voxel multi-view features and passes them to our trained Geometry Transformer to predict per-voxel material latent codes. These latents reside on a manifold of physically plausible materials, which we learn from a real-world dataset, guaranteeing the validity of decoded per-voxel materials. To obtain object-level training data, we propose an annotation pipeline combining knowledge from segmented 3D datasets, material databases, and a vision-language model, along with a new benchmark. Experiments show that VoMP estimates accurate volumetric properties, far outperforming prior art in accuracy and speed.
Authors: Yujia Zhang, Xiaoyang Wu, Yixing Lao, Chengyao Wang, Zhuotao Tian, Naiyan Wang, Hengshuang Zhao
Abstract: Humans learn abstract concepts through multisensory synergy, and once formed, such representations can often be recalled from a single modality. Inspired by this principle, we introduce Concerto, a minimalist simulation of human concept learning for spatial cognition, combining 3D intra-modal self-distillation with 2D-3D cross-modal joint embedding. Despite its simplicity, Concerto learns more coherent and informative spatial features, as demonstrated by zero-shot visualizations. It outperforms both standalone SOTA 2D and 3D self-supervised models by 14.2% and 4.8%, respectively, as well as their feature concatenation, in linear probing for 3D scene perception. With full fine-tuning, Concerto sets new SOTA results across multiple scene understanding benchmarks (e.g., 80.7% mIoU on ScanNet). We further present a variant of Concerto tailored for video-lifted point cloud spatial understanding, and a translator that linearly projects Concerto representations into CLIP's language space, enabling open-world perception. These results highlight that Concerto emerges spatial representations with superior fine-grained geometric and semantic consistency.
Authors: Yujie Wei, Shiwei Zhang, Hangjie Yuan, Yujin Han, Zhekai Chen, Jiayu Wang, Difan Zou, Xihui Liu, Yingya Zhang, Yu Liu, Hongming Shan
Abstract: Mixture-of-Experts (MoE) has emerged as a powerful paradigm for scaling model capacity while preserving computational efficiency. Despite its notable success in large language models (LLMs), existing attempts to apply MoE to Diffusion Transformers (DiTs) have yielded limited gains. We attribute this gap to fundamental differences between language and visual tokens. Language tokens are semantically dense with pronounced inter-token variation, while visual tokens exhibit spatial redundancy and functional heterogeneity, hindering expert specialization in vision MoE. To this end, we present ProMoE, an MoE framework featuring a two-step router with explicit routing guidance that promotes expert specialization. Specifically, this guidance encourages the router to partition image tokens into conditional and unconditional sets via conditional routing according to their functional roles, and refine the assignments of conditional image tokens through prototypical routing with learnable prototypes based on semantic content. Moreover, the similarity-based expert allocation in latent space enabled by prototypical routing offers a natural mechanism for incorporating explicit semantic guidance, and we validate that such guidance is crucial for vision MoE. Building on this, we propose a routing contrastive loss that explicitly enhances the prototypical routing process, promoting intra-expert coherence and inter-expert diversity. Extensive experiments on ImageNet benchmark demonstrate that ProMoE surpasses state-of-the-art methods under both Rectified Flow and DDPM training objectives. Code is available at https://github.com/ali-vilab/ProMoE.
Authors: Roman Beliy, Amit Zalcher, Jonathan Kogman, Navve Wasserman, Michal Irani
Abstract: Reconstructing images seen by people from their fMRI brain recordings provides a non-invasive window into the human brain. Despite recent progress enabled by diffusion models, current methods often lack faithfulness to the actual seen images. We present "Brain-IT", a brain-inspired approach that addresses this challenge through a Brain Interaction Transformer (BIT), allowing effective interactions between clusters of functionally-similar brain-voxels. These functional-clusters are shared by all subjects, serving as building blocks for integrating information both within and across brains. All model components are shared by all clusters & subjects, allowing efficient training with a limited amount of data. To guide the image reconstruction, BIT predicts two complementary localized patch-level image features: (i)high-level semantic features which steer the diffusion model toward the correct semantic content of the image; and (ii)low-level structural features which help to initialize the diffusion process with the correct coarse layout of the image. BIT's design enables direct flow of information from brain-voxel clusters to localized image features. Through these principles, our method achieves image reconstructions from fMRI that faithfully reconstruct the seen images, and surpass current SotA approaches both visually and by standard objective metrics. Moreover, with only 1-hour of fMRI data from a new subject, we achieve results comparable to current methods trained on full 40-hour recordings.
Authors: Zijia An, Boyu Diao, Ruiqi Liu, Libo Huang, Chuanguang Yang, Fei Wang, Zhulin An, Yongjun Xu
Abstract: Recent studies have demonstrated that incorporating trainable prompts into pretrained models enables effective incremental learning. However, the application of prompts in incremental object detection (IOD) remains underexplored. Our study reveals that existing prompt-pool-based approaches assume disjoint class sets across incremental tasks, which are unsuitable for IOD as they overlook the inherent co-occurrence phenomenon in detection. In co-occurring scenarios, unlabeled objects from previous tasks may appear in current task images, leading to confusion in prompts pool. In this paper, we hold that prompt structures should exhibit adaptive consolidation properties across tasks, with constrained updates to prevent confusion and catastrophic forgetting. Motivated by this, we introduce Parameterized Prompts for Incremental Object Detection (P$^2$IOD). Leveraging neural networks global evolution properties, P$^2$IOD employs networks as the parameterized prompts to adaptively consolidate knowledge across tasks. To constrain prompts structure updates, P$^2$IOD further engages a parameterized prompts fusion strategy. Extensive experiments on PASCAL VOC2007 and MS COCO datasets demonstrate that P$^2$IOD's effectiveness in IOD and achieves the state-of-the-art performance among existing baselines. Code is available at https://github.com/EMLS-ICTCAS/P2IOD.
Authors: Jiawei Gu, Yunzhuo Hao, Huichen Will Wang, Linjie Li, Michael Qizhe Shieh, Yejin Choi, Ranjay Krishna, Yu Cheng
Abstract: Multimodal reasoning requires iterative coordination between language and vision, yet it remains unclear what constitutes a meaningful interleaved chain of thought. We posit that text and image thoughts should function as complementary rather than isomorphic modalities that mutually advance reasoning. Guided by this principle, we build ThinkMorph, a unified model fine-tuned on approximately 24K high-quality interleaved reasoning traces spanning tasks with varying visual engagement. ThinkMorph learns to generate progressive text-image reasoning steps that concretely manipulate visual content while maintaining coherent verbal logic. It delivers large gains on vision-centric benchmarks (averaging 34.7 percent over the base model) and generalizes to out-of-domain tasks, matching or surpassing larger and proprietary VLMs. Beyond performance, ThinkMorph exhibits emergent multimodal intelligence, including unseen visual manipulation skills, adaptive switching between reasoning modes, and better test-time scaling through diversified multimodal thoughts. These findings suggest promising directions for characterizing the emergent capabilities of unified models for multimodal reasoning.
Authors: Heyu Guo, Shanmu Wang, Ruichun Ma, Shiqi Jiang, Yasaman Ghasempour, Omid Abari, Baining Guo, Lili Qiu
Abstract: Vision-language-action (VLA) models have shown strong generalization for robotic action prediction through large-scale vision-language pretraining. However, most existing models rely solely on RGB cameras, limiting their perception and, consequently, manipulation capabilities. We present OmniVLA, an omni-modality VLA model that integrates novel sensing modalities for physically-grounded spatial intelligence beyond RGB perception. The core of our approach is the sensor-masked image, a unified representation that overlays spatially grounded and physically meaningful masks onto the RGB images, derived from sensors including an infrared camera, a mmWave radar, and a microphone array. This image-native unification keeps sensor input close to RGB statistics to facilitate training, provides a uniform interface across sensor hardware, and enables data-efficient learning with lightweight per-sensor projectors. Built on this, we present a multisensory vision-language-action model architecture and train the model based on an RGB-pretrained VLA backbone. We evaluate OmniVLA on challenging real-world tasks where sensor-modality perception guides the robotic manipulation. OmniVLA achieves an average task success rate of 84%, significantly outperforms both RGB-only and raw-sensor-input baseline models by 59% and 28% respectively, meanwhile showing higher learning efficiency and stronger generalization capability.
Authors: Joonghyuk Shin, Zhengqi Li, Richard Zhang, Jun-Yan Zhu, Jaesik Park, Eli Shechtman, Xun Huang
Abstract: Current motion-conditioned video generation methods suffer from prohibitive latency (minutes per video) and non-causal processing that prevents real-time interaction. We present MotionStream, enabling sub-second latency with up to 29 FPS streaming generation on a single GPU. Our approach begins by augmenting a text-to-video model with motion control, which generates high-quality videos that adhere to the global text prompt and local motion guidance, but does not perform inference on the fly. As such, we distill this bidirectional teacher into a causal student through Self Forcing with Distribution Matching Distillation, enabling real-time streaming inference. Several key challenges arise when generating videos of long, potentially infinite time-horizons -- (1) bridging the domain gap from training on finite length and extrapolating to infinite horizons, (2) sustaining high quality by preventing error accumulation, and (3) maintaining fast inference, without incurring growth in computational cost due to increasing context windows. A key to our approach is introducing carefully designed sliding-window causal attention, combined with attention sinks. By incorporating self-rollout with attention sinks and KV cache rolling during training, we properly simulate inference-time extrapolations with a fixed context window, enabling constant-speed generation of arbitrarily long videos. Our models achieve state-of-the-art results in motion following and video quality while being two orders of magnitude faster, uniquely enabling infinite-length streaming. With MotionStream, users can paint trajectories, control cameras, or transfer motion, and see results unfold in real-time, delivering a truly interactive experience.
Authors: Haotian Xia, Haonan Ge, Junbo Zou, Hyun Woo Choi, Xuebin Zhang, Danny Suradja, Botao Rui, Ethan Tran, Wendy Jin, Zhen Ye, Xiyang Lin, Christopher Lai, Shengjie Zhang, Junwen Miao, Shichao Chen, Rhys Tracy, Vicente Ordonez, Weining Shen, Hanjie Chen
Abstract: Deeply understanding sports requires an intricate blend of fine-grained visual perception and rule-based reasoning - a challenge that pushes the limits of current multimodal models. To succeed, models must master three critical capabilities: perceiving nuanced visual details, applying abstract sport rule knowledge, and grounding that knowledge in specific visual evidence. Current sports benchmarks either cover single sports or lack the detailed reasoning chains and precise visual grounding needed to robustly evaluate these core capabilities in a multi-sport context. To address this gap, we introduce SportR, the first multi-sports large-scale benchmark designed to train and evaluate MLLMs on the fundamental reasoning required for sports intelligence. Our benchmark provides a dataset of 4,789 images and 2,052 videos. To enable granular evaluation, we structure our benchmark around a progressive hierarchy of question-answer pairs designed to probe reasoning at increasing depths - from simple infraction identification to complex penalty prediction. For the most advanced tasks requiring multi-step reasoning, such as determining penalties or explaining tactics, we provide 6,841 high-quality, human-authored Chain of Thought annotations. In addition, our benchmark incorporates both image and video modalities and provides manual bounding box annotations to test visual grounding in the image part directly. Extensive experiments demonstrate the profound difficulty of our benchmark. State-of-the-art baseline models perform poorly on our most challenging tasks. While training on our data via Supervised Fine-Tuning and Reinforcement Learning improves these scores, they remain relatively low, highlighting a significant gap in current model capabilities. SportR presents a new challenge for the community, providing a critical resource to drive future research in multimodal sports reasoning.
Authors: Feng Ding, Wenhui Yi, Yunpeng Zhou, Xinan He, Hong Rao, Shu Hu
Abstract: Fairness is a core element in the trustworthy deployment of deepfake detection models, especially in the field of digital identity security. Biases in detection models toward different demographic groups, such as gender and race, may lead to systemic misjudgments, exacerbating the digital divide and social inequities. However, current fairness-enhanced detectors often improve fairness at the cost of detection accuracy. To address this challenge, we propose a dual-mechanism collaborative optimization framework. Our proposed method innovatively integrates structural fairness decoupling and global distribution alignment: decoupling channels sensitive to demographic groups at the model architectural level, and subsequently reducing the distance between the overall sample distribution and the distributions corresponding to each demographic group at the feature level. Experimental results demonstrate that, compared with other methods, our framework improves both inter-group and intra-group fairness while maintaining overall detection accuracy across domains.
Authors: Yuetan Chu, Zhongyi Han, Gongning Luo, Xin Gao
Abstract: Understanding how segmentation performance scales with training data is fundamental for developing data-efficient medical AI systems. In this study, we systematically revisit data scaling behavior across 15 anatomical segmentation tasks spanning four imaging modalities. We observe that medical segmentation follows a structurally stable power-law-like relationship between predictive error and dataset size, characterized by rapid improvement in low-data regimes. However, unlike classical large-scale vision or language tasks, segmentation exhibits earlier and task-dependent performance saturation, with a persistent error floor emerging even as data increases. This behavior suggests that segmentation scaling is not purely data-constrained but is influenced by intrinsic geometric and anatomical structure. To further probe this geometry-constrained regime, we investigate whether topology-aware deformation-based augmentation can modify effective scaling dynamics. We compare random elastic deformation with registration-guided and generative deformation-field modeling strategies. While the overall functional form of the scaling law remains preserved, topology-aware augmentation systematically lowers the effective error scale and reshapes convergence behavior in a task-dependent manner, leading to improved sample efficiency without overturning the underlying scaling principle. These findings indicate that medical segmentation obeys a geometry-limited scaling law, and that anatomically grounded augmentation enhances data efficiency by expanding effective topological coverage rather than altering the fundamental scaling structure. Our results provide a principled empirical perspective on data-efficient learning in medical image segmentation. The code will be released after acceptance.
Authors: Jieru Lin, Zhiwei Yu, B\"orje F. Karlsson
Abstract: Autonomous agents operating in the real world must interact continuously with existing physical and semantic infrastructure, track delayed consequences, and verify outcomes over time. Everyday environments are rich in tangible control interfaces (TCIs)-e.g., light switches, appliance panels, and embedded GUI-posing core challenges for lifelong embodied agents, including partial observability, causal reasoning across time, and failure-aware verification under real-world constraints. Yet, current benchmarks rarely consider such long-horizon interaction and causality requirements. We introduce SWITCH (Semantic World Interface Tasks for Control & Handling), an embodied, task-driven benchmark created through iterative releases to probe these gaps. Its first iteration, SWITCH-Basic, evaluates five complementary abilities-task-aware VQA, semantic UI grounding, action generation, state transition prediction, and result verification-under ego-centric RGB video input and device diversity across 351 tasks spanning 98 real devices/appliances. Results from commercial and open LMMMs reveal systematic failures, highlighting critical gaps for lifelong agent deployment. SWITCH provides data, code, and held-out splits to enable reproducible non-contaminated evaluation and community contributions toward more challenging future iterations of the benchmark and the creation of relevant training data. Benchmark resources are available at: https://github.com/BAAI-Agents/SWITCH.
Authors: Zong-Wei Hong, Jing-lun Li, Lin-Ze Li, Shen Zhang, Yao Tang
Abstract: Flow Matching (FM) has recently emerged as a principled and efficient alternative to diffusion models. Standard FM encourages the learned velocity field to follow a target direction; however, it may accumulate errors along the trajectory and drive samples off the data manifold, leading to perceptual degradation, especially in lightweight or low-step configurations. To enhance stability and generalization, we extend FM into a balanced attract-repel scheme that provides explicit guidance on both "where to go" and "where not to go." To be formal, we propose \textbf{Velocity Contrastive Regularization (VeCoR)}, a complementary training scheme for flow-based generative modeling that augments the standard FM objective with contrastive, two-sided supervision. VeCoR not only aligns the predicted velocity with a stable reference direction (positive supervision) but also pushes it away from inconsistent, off-manifold directions (negative supervision). This contrastive formulation transforms FM from a purely attractive, one-sided objective into a two-sided training signal, regularizing trajectory evolution and improving perceptual fidelity across datasets and backbones. On ImageNet-1K 256$\times$256, VeCoR yields 22\% and 35\% relative FID reductions on SiT-XL/2 and REPA-SiT-XL/2 backbones, respectively, and achieves further FID gains (32\% relative) on MS-COCO text-to-image generation, demonstrating consistent improvements in stability, convergence, and image quality, particularly in low-step and lightweight settings. Project page: https://p458732.github.io/VeCoR_Project_Page/
Authors: Xinhai Hou, Shaoyuan Xu, Manan Biyani, Moyan Li, Jia Liu, Todd C. Hollon, Bryan Wang
Abstract: Agentic vision-language models are increasingly trained to "think with images" by calling image operations. However, we show that high final-answer accuracy often hides unfaithful visual reasoning: models may invoke tools on irrelevant regions or ignore tool outputs entirely, yet still guess the correct answer. In this work, we first propose a faithfulness evaluation protocol that measures whether intermediate visual tool outputs (e.g., crops) actually contain the queried evidence. This reveals that recent visual agents achieve high final-answer accuracy but exhibit low rates of faithful tool-use on visual search benchmarks. We then introduce CodeV, a code-based visual agent trained with Tool-Aware Policy Optimization (TAPO). TAPO is a process-level RL framework that augments GRPO with dense rewards defined directly on visual tool inputs and outputs, rather than on chain-of-thought tokens, making supervision easier to verify and less susceptible to reward hacking. CodeV represents visual tools as executable Python code, and TAPO assigns step-wise rewards based solely on the question and tool output, encouraging both necessary and evidence-consistent tool use. In a two-stage SFT+RL pipeline, CodeV achieves competitive or superior accuracy while substantially increasing faithful tool-use rates on related visual search benchmarks. Beyond visual search, CodeV attains strong performance on a range of multimodal reasoning and math benchmarks, suggesting that explicitly supervising intermediate tool behavior is crucial for building trustworthy, agentic visual reasoning systems.
Authors: Min Zhao, Hongzhou Zhu, Yingze Wang, Bokai Yan, Jintao Zhang, Guande He, Ling Yang, Chongxuan Li, Jun Zhu
Abstract: Despite advances, video diffusion transformers still struggle to generalize beyond their training length, a challenge we term video length extrapolation. We identify two failure modes: model-specific periodic content repetition and a universal quality degradation. Prior works attempt to solve repetition via positional encodings, overlooking quality degradation and achieving only limited extrapolation. In this paper, we revisit this challenge from a more fundamental view: attention maps, which directly govern how context influences outputs. We identify that both failure modes arise from a unified cause: attention dispersion, where tokens beyond the training window dilute learned attention patterns. This leads to quality degradation and repetition emerges as a special case when this dispersion becomes structured into periodic attention patterns, induced by harmonic properties of positional encodings. Building on this insight, we propose UltraViCo, a training-free, plug-and-play method that suppresses attention for tokens beyond the training window via a constant decay factor. By jointly addressing both failure modes, we outperform a broad set of baselines largely across models and extrapolation ratios, pushing the extrapolation limit from 2x to 4x. Remarkably, it improves Dynamic Degree and Imaging Quality by 233% and 40.5% over the previous best method at 4x extrapolation. Furthermore, our method generalizes seamlessly to downstream tasks such as controllable video synthesis and editing.
Authors: M. Naseer Subhani
Abstract: Interactive segmentation models such as the Segment Anything Model (SAM) have demonstrated remarkable generalization on natural images, but they perform suboptimally on remote sensing imagery (RSI) due to severe domain shifts and the scarcity of dense annotations. To address this limitation, we propose a point-supervised self-prompting framework that adapts SAM to RSI using only sparse point annotations. Our method employs a Refine-Requery-Reinforce loop, in which coarse pseudo-masks are generated from initial points (Refine), improved with self-constructed box prompts (Requery), and embeddings are aligned for Soft Semantic Alignment to mitigate error propagation. (Reinforce). Without relying on full-mask supervision, our approach progressively enhances SAM's segmentation quality and domain robustness through self-guided prompt adaptation. We evaluate our proposed method on three RSI benchmark datasets, WHU, HRSID, and NWPU VHR-10, showing that our method consistently surpasses pretrained SAM and recent point-supervised segmentation methods. Our results demonstrate that self-prompting and semantic alignment provide an efficient path towards scalable, point-level adaptation of foundation segmentation models for remote sensing applications.
Authors: Zichuan Lin, Yicheng Liu, Yang Yang, Lvfang Tao, Deheng Ye
Abstract: Vision-Language Models (VLMs) have achieved remarkable success in visual question answering tasks, but their reliance on large numbers of visual tokens introduces significant computational overhead. While existing efficient VLM approaches reduce visual tokens through fixed-ratio compression, they operate passively and lack the ability to adapt to varying task requirements. This motivates a fundamental question: Can VLMs autonomously determine the minimum number of visual tokens required for each sample? Inspired by human active vision mechanisms, we introduce AdaptVision, an efficient VLM paradigm that enables adaptive visual token acquisition through a coarse-to-fine approach. Our model initially processes compressed visual tokens from low-resolution images and selectively acquires additional visual information by invoking a bounding box tool to crop key regions when necessary. We train AdaptVision using a reinforcement learning framework that carefully balances accuracy and efficiency. Central to our approach is Decoupled Turn Policy Optimization (DTPO), which decouples the learning objective into two components: (1) tool learning, which optimizes correct tool utilization, and (2) accuracy improvement, which refines the generated responses to improve answer correctness. Based on this formulation, we further decouple advantage estimation by computing separate advantages for tokens associated with each objective. This formulation enables more effective optimization for AdaptVision compared to vanilla GRPO. Comprehensive experiments across multiple VQA benchmarks demonstrate that AdaptVision achieves superior performance while consuming substantially fewer visual tokens than state-of-the-art efficient VLM methods.
Authors: Hieu Dinh Trung Pham, Huy Minh Nhat Nguyen, Cuong Tuan Nguyen
Abstract: Large-scale pre-trained Vision-Language Models (VLMs) have demonstrated strong few-shot learning capabilities. However, these methods typically learn holistic representations where an image's domain-invariant structure is implicitly entangled with its domain-specific style. This presents an opportunity to further enhance generalization by disentangling these visual cues. In this paper, we propose Fourier-Attentive Representation Learning (FARL), a novel framework that addresses this by explicitly disentangling visual representations using Fourier analysis. The core of our method is a dual cross-attention mechanism, where learnable representation tokens separately query an image's structural features (from the phase spectrum) and stylistic features (from the amplitude spectrum). This process yields enriched, disentangled tokens that are then injected deep into the VLM encoders to guide adaptation. Our design, which includes an asymmetric injection strategy, forces the model to learn a more robust vision-language alignment. Extensive experiments on 15 datasets demonstrate the effectiveness of our approach.
Authors: Kuldip Singh Atwal, Dieter Pfoser, Daniel Rothbart
Abstract: Ongoing armed conflict in Sudan highlights the need for rapid monitoring of conflict-related fire-affected areas. Recent advances in deep learning and high-frequency satellite imagery enable near--real-time assessment of active fires and burn scars in war zones. This study presents a near--real-time monitoring approach using a lightweight Variational Auto-Encoder (VAE)--based model integrated with 4-band Planet Labs imagery at 3 m spatial resolution. We demonstrate that these impacted regions can be detected within approximately 24 to 30 hours under favorable observational conditions using accessible, commercially available satellite data. To achieve this, we adapt a VAE--based model, originally designed for 10-band imagery, to operate effectively on high-resolution 4-band inputs. The model is trained in an unsupervised manner to learn compact latent representations of nominal land-surface conditions and identify burn signatures by quantifying changes between temporally paired latent embeddings. Performance is evaluated across five case studies in Sudan and compared against cosine distance, CVA, and IR-MAD using precision, recall, F1-score, and the area under the precision-recall curve (AUPRC) computed between temporally paired image tiles. Results show that the proposed approach consistently outperforms the other methods, achieving higher recall and F1-scores while maintaining operationally viable precision in highly imbalanced fire-detection scenarios. Experiments with 8-band imagery and temporal image sequences yield only marginal performance gains over single 4-band inputs, underscoring the effectiveness of the proposed lightweight approach for scalable, near--real-time conflict monitoring.
Authors: Chandler Timm C. Doloriel, Yunbei Zhang, Yeonguk Yu, Taki Hasan Rafi, Muhammad salman siddiqui, Tor Kristian Stevik, Habib Ullah, Fadi Al Machot, Kristian Hovde Liland
Abstract: Recent continual test-time adaptation (CTTA) methods adopt masked image modeling to stabilize learning under distribution shift, yet each treats its masking family $F$ as a fixed design choice and innovates exclusively along the selection strategy $S$, leaving the family axis underexplored. We present a systematic empirical study that isolates this axis. Using a controlled CTTA instantiation -- Mask to Adapt (M2A) -- that fixes $S=random$ and standard losses, we vary only $F$ across spatial (patch, pixel) and frequency (all-band, low-band, high-band) families while keeping every other component identical. The study's contributions are the design guidance it extracts for the CTTA settings we evaluated: (1)~\emph{the masking family determines whether adaptation compounds useful structure or compounds errors} -- on patch-tokenized architectures, spatial masking accumulates stable representations over long streams while frequency masking collapses catastrophically. We characterize this instability through a \emph{structural-preservation} account, where spatial coherence maintains the broad-spectrum redundancy needed to avoid terminally overlapping with a corruption's spectral signature; (2)~\emph{the optimal family depends on architecture-task alignment} -- on CNNs, whose overlapping receptive fields dilute patch occlusion, the family gap vanishes, whereas on fine-grained tasks with global cues and large-capacity ViTs, frequency masking becomes competitive. In confounded system-level comparisons -- where baselines also differ in losses and auxiliary components -- M2A's random selection performs comparably to heuristic strategies, though we treat this observation as suggestive context rather than a controlled quantification of $S$'s relative importance.
Authors: Romain Seailles (DI-ENS, WILLOW), Jean-Baptiste Masson (EPIMETHEE), Jean Ponce (DI-ENS, CDS, WILLOW), Julien Mairal (Thoth)
Abstract: Single-molecule localization microscopy (SMLM) allows reconstructing biology-relevant structures beyond the diffraction limit by detecting and localizing individual fluorophores -- fluorescent molecules stained onto the observed specimen -- over time to reconstruct super-resolved images. Currently, efficient SMLM requires non-overlapping emitting fluorophores, leading to long acquisition times that hinders live-cell imaging. Recent deep-learning approaches can handle denser emissions, but they rely on variants of non-maximum suppression (NMS) layers, which are unfortunately non-differentiable and may discard true positives with their local fusion strategy. In this presentation, we reformulate the SMLM training objective as a set-matching problem, deriving an optimal-transport loss that eliminates the need for NMS during inference and enables end-to-end training. Additionally, we propose an iterative neural network that integrates knowledge of the microscope's optical system inside our model. Experiments on synthetic benchmarks and real biological data show that both our new loss function and architecture surpass the state of the art at moderate and high emitter densities. Code is available at https://github.com/RSLLES/SHOT.
Authors: Fatimah Zohra, Chen Zhao, Hani Itani, Bernard Ghanem
Abstract: CLIP achieves strong zero-shot image-text retrieval by aligning global vision and text representations, yet it falls behind on fine-grained tasks even when fine-tuned on long, detailed captions. In this work, we propose $\beta$-CLIP, a multi-granular text-conditioned contrastive learning framework designed to achieve hierarchical alignment between multiple textual granularities-from full captions to sentences and phrases-and their corresponding visual regions. For each level of granularity, $\beta$-CLIP utilizes cross-attention to dynamically pool image patches, producing contextualized visual embeddings. To address the semantic overlap inherent in this hierarchy, we introduce the $\beta$-Contextualized Contrastive Alignment Loss ($\beta$-CAL). This objective parameterizes the trade-off between strict query-specific matching and relaxed intra-image contextualization, supporting both soft Cross-Entropy and hard Binary Cross-Entropy formulations. We find that each loss interacts differently with hierarchical supervision: CE's softmax sharpens fine-grained discrimination, while BCE's sigmoid favors long-text retrieval while both benefit from hierarchy. Through extensive experiments, we demonstrate that $\beta$-CLIP significantly improves dense alignment: achieving 91.8% T2I 92.3% I2T at R@1 on Urban1K and 30.9% on FG-OVD (Hard), setting state-of-the-art among methods trained without hard negatives. $\beta$-CLIP establishes a robust, adaptive baseline for dense vision-language correspondence. The code and models are released at https://github.com/fzohra/B-CLIP.
Authors: Jie Zhang, Shuai Dong, Shiguang Shan, Xilin Chen
Abstract: Recent approaches employing imperceptible perturbations in input images have demonstrated promising potential to counter malicious manipulations in diffusion-based image editing systems. However, existing methods suffer from limited transferability in cross-model evaluations. To address this, we propose Transferable Defense Against Malicious Image Edits (TDAE), a novel bimodal framework that enhances image immunity against malicious edits through coordinated image-text optimization. Specifically, at the visual defense level, we introduce FlatGrad Defense Mechanism (FDM), which incorporates gradient regularization into the adversarial objective. By explicitly steering the perturbations toward flat minima, FDM amplifies immune robustness against unseen editing models. For textual enhancement protection, we propose an adversarial optimization paradigm named Dynamic Prompt Defense (DPD), which periodically refines text embeddings to align the editing outcomes of immunized images with those of the original images, then updates the images under optimized embeddings. Through iterative adversarial updates to diverse embeddings, DPD enforces the generation of immunized images that seek a broader set of immunity-enhancing features, thereby achieving cross-model transferability. Extensive experimental results demonstrate that our TDAE achieves state-of-the-art performance in mitigating malicious edits under both intra- and cross-model evaluations.
Authors: Zihan Wang, Jiashun Wang, Jeff Tan, Yiwen Zhao, Jessica Hodgins, Shubham Tulsiani, Deva Ramanan
Abstract: We introduce CRISP, a method that recovers simulatable human motion and scene geometry from monocular video. Prior work on joint human-scene reconstruction relies on data-driven priors and joint optimization with no physics in the loop, or recovers noisy geometry with artifacts that cause motion tracking policies with scene interactions to fail. In contrast, our key insight is to recover convex, clean, and simulation-ready geometry by fitting planar primitives to a point cloud reconstruction of the scene, via a simple clustering pipeline over depth, normals, and flow. To reconstruct scene geometry that might be occluded during interactions, we make use of human-scene contact modeling (e.g., we use human posture to reconstruct the occluded seat of a chair). Finally, we ensure that human and scene reconstructions are physically-plausible by using them to drive a humanoid controller via reinforcement learning. Our approach reduces motion tracking failure rates from 55.2\% to 6.9\% on human-centric video benchmarks (EMDB, PROX), while delivering a 43\% faster RL simulation throughput. We further validate it on in-the-wild videos including casually-captured videos, Internet videos, and even Sora-generated videos. This demonstrates CRISP's ability to generate physically-valid human motion and interaction environments at scale, greatly advancing real-to-sim applications for robotics and AR/VR.
Authors: Jingyu Li, Xiaolong Zhao, Zhe Liu, Wenxiao Wu, Li Zhang
Abstract: Semi-supervised 3D object detection, aiming to explore unlabeled data for boosting 3D object detectors, has emerged as an active research area in recent years. Some previous methods have shown substantial improvements by either employing heterogeneous teacher models to provide high-quality pseudo labels or enforcing feature-perspective consistency between the teacher and student networks. However, these methods overlook the fact that the model usually tends to exhibit low sensitivity to object geometries with limited labeled data, making it difficult to capture geometric information, which is crucial for enhancing the student model's ability in object perception and localization. In this paper, we propose GeoTeacher to enhance the student model's ability to capture geometric relations of objects with limited training data, especially unlabeled data. We design a keypoint-based geometric relation supervision module that transfers the teacher model's knowledge of object geometry to the student, thereby improving the student's capability in understanding geometric relations. Furthermore, we introduce a voxel-wise data augmentation strategy that increases the diversity of object geometries, thereby further improving the student model's ability to comprehend geometric structures. To preserve the integrity of distant objects during augmentation, we incorporate a distance-decay mechanism into this strategy. Moreover, GeoTeacher can be combined with different SS3D methods to further improve their performance. Extensive experiments on the ONCE and Waymo datasets indicate the effectiveness and generalization of our method and we achieve the new state-of-the-art results. Code will be available at https://github.com/SII-Whaleice/GeoTeacher
Authors: Tong Shao, Yusen Fu, Guoying Sun, Jingde Kong, Zhuotao Tian, Jingyong Su
Abstract: Although Diffusion Transformer (DiT) has emerged as a predominant architecture for image and video generation, its iterative denoising process results in slow inference, which hinders broader applicability and development. Caching-based methods achieve training-free acceleration, while suffering from considerable computational error. Existing methods typically incorporate error correction strategies such as pruning or prediction to mitigate it. However, their fixed caching strategy fails to adapt to the complex error variations during denoising, which limits the full potential of error correction. To tackle this challenge, we propose a novel fidelity-optimization plugin for existing error correction methods via cumulative error minimization, named CEM. CEM predefines the error to characterize the sensitivity of model to acceleration jointly influenced by timesteps and cache intervals. Guided by this prior, we formulate a dynamic programming algorithm with cumulative error approximation for strategy optimization, which achieves the caching error minimization, resulting in a substantial improvement in generation fidelity. CEM is model-agnostic and exhibits strong generalization, which is adaptable to arbitrary acceleration budgets. It can be seamlessly integrated into existing error correction frameworks and quantized models without introducing any additional computational overhead. Extensive experiments conducted on nine generation models and quantized methods across three tasks demonstrate that CEM significantly improves generation fidelity of existing acceleration models, and outperforms the original generation performance on FLUX.1-dev, PixArt-$\alpha$, StableDiffusion1.5 and Hunyuan. Our code is released publicly at https://github.com/leaves162/CEM.
Authors: Xin Jin, Yichuan Zhong, Yapeng Tian
Abstract: Current text-conditioned diffusion editors handle single object replacement well but struggle when a new object and a new style must be introduced simultaneously. We present Twin-Prompt Attention Blend (TP-Blend), a lightweight training-free framework that receives two separate textual prompts, one specifying a blend object and the other defining a target style, and injects both into a single denoising trajectory. TP-Blend is driven by two complementary attention processors. Cross-Attention Object Fusion (CAOF) first averages head-wise attention to locate spatial tokens that respond strongly to either prompt, then solves an entropy-regularised optimal transport problem that reassigns complete multi-head feature vectors to those positions. CAOF updates feature vectors at the full combined dimensionality of all heads (e.g., 640 dimensions in SD-XL), preserving rich cross-head correlations while keeping memory low. Self-Attention Style Fusion (SASF) injects style at every self-attention layer through Detail-Sensitive Instance Normalization. A lightweight one-dimensional Gaussian filter separates low- and high-frequency components; only the high-frequency residual is blended back, imprinting brush-stroke-level texture without disrupting global geometry. SASF further swaps the Key and Value matrices with those derived from the style prompt, enforcing context-aware texture modulation that remains independent of object fusion. Extensive experiments show that TP-Blend produces high-resolution, photo-realistic edits with precise control over both content and appearance, surpassing recent baselines in quantitative fidelity, perceptual quality, and inference speed.
Authors: Yujian Lee, Peng Gao, Yongqi Xu, Wentao Fan
Abstract: Audio-visual semantic segmentation (AVSS) represents an extension of the audio-visual segmentation (AVS) task, necessitating a semantic understanding of audio-visual scenes beyond merely identifying sound-emitting objects at the visual pixel level. Contrary to a previous methodology, by decomposing the AVSS task into two discrete subtasks by initially providing a prompted segmentation mask to facilitate subsequent semantic analysis, our approach innovates on this foundational strategy. We introduce a novel collaborative framework, \textit{S}tepping \textit{S}tone \textit{P}lus (SSP), which integrates optical flow and textual prompts to assist the segmentation process. In scenarios where sound sources frequently coexist with moving objects, our pre-mask technique leverages optical flow to capture motion dynamics, providing essential temporal context for precise segmentation. To address the challenge posed by stationary sound-emitting objects, such as alarm clocks, SSP incorporates two specific textual prompts: one identifies the category of the sound-emitting object, and the other provides a broader description of the scene. Additionally, we implement a visual-textual alignment module (VTA) to facilitate cross-modal integration, delivering more coherent and contextually relevant semantic interpretations. Our training regimen involves a post-mask technique aimed at compelling the model to learn the diagram of the optical flow. Experimental results demonstrate that SSP outperforms existing AVS methods, delivering efficient and precise segmentation results.
Authors: Wenxuan Huang, Yu Zeng, Qiuchen Wang, Zhen Fang, Shaosheng Cao, Zheng Chu, Qingyu Yin, Shuang Chen, Zhenfei Yin, Lin Chen, Zehui Chen, Xu Tang, Yao Hu, Philip Torr, Feng Zhao, Wanli Ouyang
Abstract: Multimodal large language models (MLLMs) have achieved remarkable success across a broad range of vision tasks. However, constrained by the capacity of their internal world knowledge, prior work has proposed augmenting MLLMs by ``reasoning-then-tool-call'' for visual and textual search engines to obtain substantial gains on tasks requiring extensive factual information. However, these approaches typically define multimodal search in a naive setting, assuming that a single full-level or entity-level image query and few text query suffices to retrieve the key evidence needed to answer the question, which is unrealistic in real-world scenarios with substantial visual noise. Moreover, they are often limited in the reasoning depth and search breadth, making it difficult to solve complex questions that require aggregating evidence from diverse visual and textual sources. Building on this, we propose Vision-DeepResearch, which proposes one new multimodal deep-research paradigm, i.e., performs multi-turn, multi-entity and multi-scale visual and textual search to robustly hit real-world search engines under heavy noise. Our Vision-DeepResearch supports dozens of reasoning steps and hundreds of engine interactions, while internalizing deep-research capabilities into the MLLM via cold-start supervision and RL training, resulting in a strong end-to-end multimodal deep-research MLLM. It substantially outperforming existing multimodal deep-research MLLMs, and workflows built on strong closed-source foundation model such as GPT-5, Gemini-2.5-pro and Claude-4-Sonnet. The code will be released in https://github.com/Osilly/Vision-DeepResearch.
Authors: Shashank Mishra, Didier Stricker, Jason Rambach
Abstract: Anomaly detection is often formulated under the assumption that abnormality is an intrinsic property of an observation, independent of context. This assumption breaks down in many real-world settings, where the same object or action may be normal or anomalous depending on latent contextual factors (e.g., running on a track versus on a highway). We revisit \emph{contextual anomaly detection}, classically defined as context-dependent abnormality, and operationalize it in the visual domain, where anomaly labels depend on subject--context compatibility rather than intrinsic appearance. To enable systematic study of this setting, we introduce CAAD-3K, a benchmark that isolates contextual anomalies by controlling subject identity while varying context. We further propose a conditional compatibility learning framework that leverages vision--language representations to model subject--context relationships under limited supervision. Our method substantially outperforms existing approaches on CAAD-3K and achieves state-of-the-art performance on MVTec-AD and VisA, demonstrating that modeling context dependence complements traditional structural anomaly detection. Our code and dataset will be publicly released.
Authors: Hari Krishna Gadi, Daniel Matos, Hongyi Luo, Lu Liu, Yongliang Wang, Yanfeng Zhang, Liqiu Meng
Abstract: Visual geolocalization, the task of predicting where an image was taken, remains challenging due to global scale, visual ambiguity, and the inherently hierarchical structure of geography. Existing paradigms rely on either large-scale retrieval, which requires storing a large number of image embeddings, grid-based classifiers that ignore geographic continuity, or generative models that diffuse over space but struggle with fine detail. We introduce an entity-centric formulation of geolocation that replaces image-to-image retrieval with a compact hierarchy of geographic entities embedded in Hyperbolic space. Images are aligned directly to country, region, subregion, and city entities through Geo-Weighted Hyperbolic contrastive learning by directly incorporating haversine distance into the contrastive objective. This hierarchical design enables interpretable predictions and efficient inference with 240k entity embeddings instead of over 5 million image embeddings on the OSV5M benchmark, on which our method establishes a new state-of-the-art performance. Compared to the current methods in the literature, it reduces mean geodesic error by 19.5\%, while improving the fine-grained subregion accuracy by 43%. These results demonstrate that geometry-aware hierarchical embeddings provide a scalable and conceptually new alternative for global image geolocation.
Authors: Meng Luo, Bobo Li, Shanqing Xu, Shize Zhang, Qiuchan Chen, Menglu Han, Wenhao Chen, Yanxiang Huang, Hao Fei, Mong-Li Lee, Wynne Hsu
Abstract: Despite rapid progress in multimodal large language models (MLLMs), their capability for deep emotional understanding remains limited. We argue that genuine affective intelligence requires explicit modeling of Theory of Mind (ToM), the cognitive substrate from which emotions arise. To this end, we introduce HitEmotion, a ToM-grounded hierarchical benchmark that diagnoses capability breakpoints across increasing levels of cognitive depth. Second, we propose a ToM-guided reasoning chain that tracks mental states and calibrates cross-modal evidence to achieve faithful emotional reasoning. We further introduce TMPO, a reinforcement learning method that uses intermediate mental states as process-level supervision to guide and strengthen model reasoning. Extensive experiments show that HitEmotion exposes deep emotional reasoning deficits in state-of-the-art models, especially on cognitively demanding tasks. In evaluation, the ToM-guided reasoning chain and TMPO improve end-task accuracy and yield more faithful, more coherent rationales. In conclusion, our work provides the research community with a practical toolkit for evaluating and enhancing the cognition-based emotional understanding capabilities of MLLMs. Our dataset and code are available at: https://HitEmotion.github.io/.
Authors: Yinchao Ma, Qiang Zhou, Zhibin Wang, Xianing Chen, Hanqing Yang, Jun Song, Bo Zheng
Abstract: Video large language models have demonstrated remarkable capabilities in video understanding tasks. However, the redundancy of video tokens introduces significant computational overhead during inference, limiting their practical deployment. Many compression algorithms are proposed to prioritize retaining features with the highest attention scores to minimize perturbations in attention computations. However, the correlation between attention scores and their actual contribution to correct answers remains ambiguous. To address the above limitation, we propose a novel \textbf{C}ontribution-\textbf{a}ware token \textbf{Co}mpression algorithm for \textbf{VID}eo understanding (\textbf{CaCoVID}) that explicitly optimizes the token selection policy based on the contribution of tokens to correct predictions. First, we introduce a reinforcement learning-based framework that optimizes a policy network to select video token combinations with the greatest contribution to correct predictions. This paradigm shifts the focus from passive token preservation to active discovery of optimal compressed token combinations. Secondly, we propose a combinatorial policy optimization algorithm with online combination space sampling, which dramatically reduces the exploration space for video token combinations and accelerates the convergence speed of policy optimization. Extensive experiments on diverse video understanding benchmarks demonstrate the effectiveness of CaCoVID. Codes are available at https://github.com/LivingFutureLab/CaCoVID.
Authors: Shicheng Yin, Kaixuan Yin, Weixing Chen, Yang Liu, Guanbin Li, Liang Lin
Abstract: World models are essential for autonomous robotic planning. However, the substantial computational overhead of existing dense Transformerbased models significantly hinders real-time deployment. To address this efficiency-performance bottleneck, we introduce DDP-WM, a novel world model centered on the principle of Disentangled Dynamics Prediction (DDP). We hypothesize that latent state evolution in observed scenes is heterogeneous and can be decomposed into sparse primary dynamics driven by physical interactions and secondary context-driven background updates. DDP-WM realizes this decomposition through an architecture that integrates efficient historical processing with dynamic localization to isolate primary dynamics. By employing a crossattention mechanism for background updates, the framework optimizes resource allocation and provides a smooth optimization landscape for planners. Extensive experiments demonstrate that DDP-WM achieves significant efficiency and performance across diverse tasks, including navigation, precise tabletop manipulation, and complex deformable or multi-body interactions. Specifically, on the challenging Push-T task, DDP-WM achieves an approximately 9 times inference speedup and improves the MPC success rate from 90% to98% compared to state-of-the-art dense models. The results establish a promising path for developing efficient, high-fidelity world models. Codes will be available at https://github.com/HCPLab-SYSU/DDP-WM.
Authors: Yuliang Zhan, Jian Li, Wenbing Huang, Wenbing Huang, Yang Liu, Hao Sun
Abstract: Deep learning has demonstrated remarkable capabilities in simulating complex dynamic systems. However, existing methods require known physical properties as supervision or inputs, limiting their applicability under unknown conditions. To explore this challenge, we introduce Cloth Dynamics Grounding (CDG), a novel scenario for unsupervised learning of cloth dynamics from multi-view visual observations. We further propose Cloth Dynamics Splatting (CloDS), an unsupervised dynamic learning framework designed for CDG. CloDS adopts a three-stage pipeline that first performs video-to-geometry grounding and then trains a dynamics model on the grounded meshes. To cope with large non-linear deformations and severe self-occlusions during grounding, we introduce a dual-position opacity modulation that supports bidirectional mapping between 2D observations and 3D geometry via mesh-based Gaussian splatting in video-to-geometry grounding stage. It jointly considers the absolute and relative position of Gaussian components. Comprehensive experimental evaluations demonstrate that CloDS effectively learns cloth dynamics from visual data while maintaining strong generalization capabilities for unseen configurations. Our code is available at https://github.com/whynot-zyl/CloDS. Visualization results are available at https://github.com/whynot-zyl/CloDS_video}.%\footnote{As in this example.
URLs: https://github.com/whynot-zyl/CloDS., https://github.com/whynot-zyl/CloDS_video
Authors: Yu Zeng, Wenxuan Huang, Zhen Fang, Shuang Chen, Yufan Shen, Yishuo Cai, Xiaoman Wang, Zhenfei Yin, Lin Chen, Zehui Chen, Shiting Huang, Yiming Zhao, Xu Tang, Yao Hu, Philip Torr, Wanli Ouyang, Shaosheng Cao
Abstract: Multimodal Large Language Models (MLLMs) have advanced VQA and now support Vision-DeepResearch systems that use search engines for complex visual-textual fact-finding. However, evaluating these visual and textual search abilities is still difficult, and existing benchmarks have two major limitations. First, existing benchmarks are not visual search-centric: answers that should require visual search are often leaked through cross-textual cues in the text questions or can be inferred from the prior world knowledge in current MLLMs. Second, overly idealized evaluation scenario: On the image-search side, the required information can often be obtained via near-exact matching against the full image, while the text-search side is overly direct and insufficiently challenging. To address these issues, we construct the Vision-DeepResearch benchmark (VDR-Bench) comprising 2,000 VQA instances. All questions are created via a careful, multi-stage curation pipeline and rigorous expert review, designed to assess the behavior of Vision-DeepResearch systems under realistic real-world conditions. Moreover, to address the insufficient visual retrieval capabilities of current MLLMs, we propose a simple multi-round cropped-search workflow. This strategy is shown to effectively improve model performance in realistic visual retrieval scenarios. Overall, our results provide practical guidance for the design of future multimodal deep-research systems. The code will be released in https://github.com/Osilly/Vision-DeepResearch.
Authors: Wangduo Xie, Matthew B. Blaschko
Abstract: Computed Tomography (CT) plays a vital role in inspecting the internal structures of industrial objects. Furthermore, achieving high-quality CT reconstruction from sparse views is essential for reducing production costs. While classic implicit neural networks have shown promising results for sparse reconstruction, they are unable to leverage shape priors of objects. Motivated by the observation that numerous industrial objects exhibit rectangular structures, we propose a novel Neural Adaptive Binning (NAB) method that effectively integrates rectangular priors into the reconstruction process. Specifically, our approach first maps coordinate space into a binned vector space. This mapping relies on an innovative binning mechanism based on differences between shifted hyperbolic tangent functions, with our extension enabling rotations around the input-plane normal vector. The resulting representations are then processed by a neural network to predict CT attenuation coefficients. This design enables end-to-end optimization of the encoding parameters -- including position, size, steepness, and rotation -- via gradient flow from the projection data, thus enhancing reconstruction accuracy. By adjusting the smoothness of the binning function, NAB can generalize to objects with more complex geometries. This research provides a new perspective on integrating shape priors into neural network-based reconstruction. Extensive experiments demonstrate that NAB achieves superior performance on two industrial datasets. It also maintains robust on medical datasets when the binning function is extended to more general expression. The code is available at https://github.com/Wangduo-Xie/NAB_CT_reconstruction.
Authors: Mohammadreza Salehi, Mehdi Noroozi, Luca Morreale, Ruchika Chavhan, Malcolm Chadwick, Alberto Gil Ramos, Abhinav Mehrotra
Abstract: Instructional video editing applies edits to an input video using only text prompts, enabling intuitive natural-language control. Despite rapid progress, most methods still require fixed-length inputs and substantial compute. Meanwhile, autoregressive video generation enables efficient variable-length synthesis, yet remains under-explored for video editing. We introduce a causal, efficient video editing model that edits variable-length videos frame by frame. For efficiency, we start from a 2D image-to-image (I2I) diffusion model and adapt it to video-to-video (V2V) editing by conditioning the edit at time step t on the model's prediction at t-1. To leverage videos' temporal redundancy, we propose a new I2I diffusion forward process formulation that encourages the model to predict the residual between the target output and the previous prediction. We call this Residual Flow Diffusion Model (RFDM), which focuses the denoising process on changes between consecutive frames. Moreover, we propose a new benchmark that better ranks state-of-the-art methods for editing tasks. Trained on paired video data for global/local style transfer and object removal, RFDM surpasses I2I-based methods and competes with fully spatiotemporal (3D) V2V models, while matching the compute of image models and scaling independently of input video length. More content can be found in: https://smsd75.github.io/RFDM_page/
Authors: Yan Luo, Advaith Ravishankar, Serena Liu, Yutong Yang, Mengyu Wang
Abstract: While three-dimensional imaging is essential for clinical diagnosis, its high cost and long wait times have motivated the use of image-to-3D foundation models to infer volume from two-dimensional modalities. However, because these models are trained on natural images, their learned geometric priors struggle to transfer to inherently planar medical data. A benchmark of five state-of-the-art models (SAM3D, Hunyuan3D-2.1, Direct3D, Hi3DGen, and TripoSG) across six medical and two natural datasets revealed that voxel-based overlap remains uniformly low across all methods due to severe depth ambiguity from single-slice inputs. Despite this fundamental volumetric failure, global distance metrics indicate that SAM3D best captures topological similarity to ground-truth medical shapes, whereas alternative models are prone to oversimplification. Ultimately, these findings quantify the limits of zero-shot single-slice 3D inference, highlighting that reliable medical 3D reconstruction requires domain-specific adaptation and anatomical constraints to overcome complex medical geometries.
Authors: Zesheng Jia, Jin Wang, Siao Liu, Lingzhi Li, Ziyao Huang, Yunjiang Xu, Jianping Wang
Abstract: Fast domain adaptation remains a fundamental challenge for deploying multi-agent systems across diverse environments in Vehicle-to-Everything (V2X) collaborative perception. Despite the success of Parameter-Efficient Fine-Tuning (PEFT) in natural language processing and conventional vision tasks, directly applying PEFT to multi-agent settings leads to significant performance degradation and training instability. In this work, we conduct a detailed analysis and identify two key factors: (i) inter-frame redundancy in heterogeneous sensory streams, and (ii) erosion of fine-grained semantics in deep-layer representations under PEFT adaptation. To address these issues, we propose FlowAdapt, a parameter-efficient framework grounded in optimal transport theory, which minimizes information transport costs across both data distributions and network hierarchies. Specifically, we introduce a Wasserstein Greedy Sampling strategy to selectively filter redundant samples via a bounded covering radius. Furthermore, Progressive Knowledge Transfer module is designed to progressively inject compressed early-stage representations into later stages through learnable pathways, alleviating semantic degradation in late-stage adaptation. Extensive experiments on three benchmarks demonstrate that FlowAdapt achieves state-of-the-art performance with only 1% of trainable parameters, effectively bridging domain gaps with superior sample efficiency and generalization.
Authors: Rang Meng, Yingjie Yin, Yuming Li, Chenguang Ma
Abstract: Recent multi-modal video generation models have achieved high visual quality, but their prohibitive latency and limited temporal stability hinder real-time deployment. Streaming inference exacerbates these issues, leading to pronounced multimodal degradation, such as spatial blurring, temporal drift, and lip desynchronization, which creates an unresolved efficiency-performance trade-off. To this end, we propose EchoTorrent, a novel schema with a fourfold design: (1) Multi-Teacher Training fine-tunes a pre-trained model on distinct preference domains to obtain specialized domain experts, which sequentially transfer domain-specific knowledge to a student model; (2) Adaptive CFG Calibration (ACC-DMD), which calibrates the audio CFG augmentation errors in DMD via a phased spatiotemporal schedule, eliminating redundant CFG computations and enabling single-pass inference per step; (3) Hybrid Long Tail Forcing, which enforces alignment exclusively on tail frames during long-horizon self-rollout training via a causal-bidirectional hybrid architecture, effectively mitigates spatiotemporal degradation in streaming mode while enhancing fidelity to reference frames; and (4) VAE Decoder Refiner through pixel-domain optimization of the VAE decoder to recover high-frequency details while circumventing latent-space ambiguities. Extensive experiments and analysis demonstrate that EchoTorrent achieves few-pass autoregressive generation with substantially extended temporal consistency, identity preservation, and audio-lip synchronization.
Authors: Shaobin Zhuang, Yuang Ai, Jiaming Han, Weijia Mao, Xiaohui Li, Fangyikang Wang, Xiao Wang, Yan Li, Shanchuan Lin, Kun Xu, Zhenheng Yang, Huaibo Huang, Xiangyu Yue, Hao Chen, Yali Wang
Abstract: Unified Multimodal Large Language Models (MLLMs) require a visual representation that simultaneously supports high-fidelity reconstruction, complex semantic extraction, and generative suitability. However, existing visual tokenizers typically struggle to satisfy these conflicting objectives within a single framework. In this paper, we introduce UniWeTok, a unified discrete tokenizer designed to bridge this gap using a massive binary codebook ($\mathit{2^{128}}$). For training framework, we introduce Pre-Post Distillation and a Generative-Aware Prior to enhance the semantic extraction and generative prior of the discrete tokens. In terms of model architecture, we propose a convolution-attention hybrid architecture with the SigLu activation function. SigLu activation not only bounds the encoder output and stabilizes the semantic distillation process but also effectively addresses the optimization conflict between token entropy loss and commitment loss. We further propose a three-stage training framework designed to enhance UniWeTok's adaptability cross various image resolutions and perception-sensitive scenarios, such as those involving human faces and textual content. On ImageNet, UniWeTok achieves state-of-the-art image generation performance (FID: UniWeTok 1.38 vs. REPA 1.42) while requiring a remarkably low training compute (Training Tokens: UniWeTok 33B vs. REPA 262B). On general-domain, UniWeTok demonstrates highly competitive capabilities across a broad range of tasks, including multimodal understanding, image generation (DPG Score: UniWeTok 86.63 vs. FLUX.1 [Dev] 83.84), and editing (GEdit Overall Score: UniWeTok 5.09 vs. OmniGen 5.06). We release code and models to facilitate community exploration of unified tokenizer and MLLM.
Authors: Yiwen Wang, Jiahao Qin
Abstract: We address the problem of cross-domain image registration, where paired images exhibit coupled geometric misalignment and domain-specific appearance shift. We formalize this as a factorization problem: decomposing each image into a domain-invariant scene representation and a global appearance statistic, such that registration reduces to recombining the scene structure of the moving image with the appearance of the fixed image via Adaptive Instance Normalization (AdaIN). This factorization eliminates the need for explicit deformation field estimation. To exploit temporal coherence in sequential acquisitions, we introduce a position-encoded cross-frame attention mechanism that fuses learnable and sinusoidal position embeddings with multi-head attention over a sliding window of neighboring frames, enriching the scene representation with inter-frame context. We instantiate this framework as GPEReg-Net and evaluate on two benchmarks: FIRE-Reg-256 (retinal fundus, semi-rigid) and HPatches-Reg-256 (synthetic textured patches, affine). GPEReg-Net achieves state-of-the-art performance on both benchmarks (FIRE: SSIM = 0.928, PSNR = 33.47 dB; HPatches: SSIM = 0.450, PSNR = 21.01 dB), surpassing all baselines, including deformation-based methods, while running 1.87x faster than SAS-Net. Code: https://github.com/JiahaoQin/GPEReg-Net.
Authors: Tianwei Lin, Zhongwei Qiu, Wenqiao Zhang, Jiang Liu, Yihan Xie, Mingjian Gao, Zhenxuan Fan, Zhaocheng Li, Sijing Li, Zhongle Xie, Peng LU, Yueting Zhuang, Ling Zhang, Beng Chin Ooi, Yingda Xia
Abstract: Computed Tomography (CT) is one of the most widely used and diagnostically information-dense imaging modalities, covering critical organs such as the heart, lungs, liver, and colon. Clinical interpretation relies on both slice-driven local features (e.g., sub-centimeter nodules, lesion boundaries) and volume-driven spatial representations (e.g., tumor infiltration, inter-organ anatomical relations). However, existing Large Vision-Language Models (LVLMs) remain fragmented in CT slice versus volumetric understanding: slice-driven LVLMs show strong generalization but lack cross-slice spatial consistency, while volume-driven LVLMs explicitly capture volumetric semantics but suffer from coarse granularity and poor compatibility with slice inputs. The absence of a unified modeling paradigm constitutes a major bottleneck for the clinical translation of medical LVLMs. We present OmniCT, a powerful unified slice-volume LVLM for CT scenarios, which makes three contributions: (i) Spatial Consistency Enhancement (SCE): volumetric slice composition combined with tri-axial positional embedding that introduces volumetric consistency, and an MoE hybrid projection enables efficient slice-volume adaptation; (ii) Organ-level Semantic Enhancement (OSE): segmentation and ROI localization explicitly align anatomical regions, emphasizing lesion- and organ-level semantics; (iii) MedEval-CT: the largest slice-volume CT dataset and hybrid benchmark integrates comprehensive metrics for unified evaluation. OmniCT consistently outperforms existing methods with a substantial margin across diverse clinical tasks and satisfies both micro-level detail sensitivity and macro-level spatial reasoning. More importantly, it establishes a new paradigm for cross-modal medical imaging understanding. Our project is available at https://github.com/ZJU4HealthCare/OmniCT.
Authors: Bo Liang, Chen Gong, Haobo Wang, Qirui Liu, Rungui Zhou, Fengzhi Shao, Yubo Wang, Wei Gao, Kaichen Zhou, Guolong Cui, Chenren Xu
Abstract: Millimeter-wave Human Pose Estimation (mmWave HPE) promises privacy but suffers from poor generalization under distribution shifts. We demonstrate that brute-force data scaling is ineffective for out-of-distribution (OOD) robustness; efficiency and coverage are the true bottlenecks. To address this, we introduce WiCompass, a coverage-aware data-collection framework. WiCompass leverages large-scale motion-capture corpora to build a universal pose space ``oracle'' that quantifies dataset redundancy and identifies underrepresented motions. Guided by this oracle, WiCompass employs a closed-loop policy to prioritize collecting informative missing samples. Experiments show that WiCompass consistently improves OOD accuracy at matched budgets and exhibits superior scaling behavior compared to conventional collection strategies. By shifting focus from brute-force scaling to coverage-aware data acquisition, this work offers a practical path toward robust mmWave sensing.
Authors: Miaowei Wang, Qingxuan Yan, Zhi Cao, Yayuan Li, Oisin Mac Aodha, Jason J. Corso, Amir Vaxman
Abstract: Text-guided dynamic 3D character generation has advanced rapidly, yet producing high-quality motion that faithfully reflects rich textual descriptions remains challenging. Existing methods tend to generate limited sub-actions or incoherent motion due to fixed-length temporal inputs and discrete frame-wise representations that fail to capture rich motion semantics. We address these limitations by representing motion with continuous differentiable B-spline curves, enabling more effective motion generation without modifying the capabilities of the underlying generative model. Specifically, our closed-form, Laplacian-regularized B-spline solver efficiently compresses variable-length motion sequences into compact representations with a fixed number of control points. Further, we introduce a normal-fusion strategy for input shape adherence along with correspondence-aware and local-rigidity losses for motion-restoration quality. To train our model, we collate BIMO, a new dataset containing diverse variable-length 3D motion sequences with rich, high-quality text annotations. Extensive evaluations show that our feed-forward framework BiMotion generates more expressive, higher-quality, and better prompt-aligned motions than existing state-of-the-art methods, while also achieving faster generation. Our project page is at: https://wangmiaowei.github.io/BiMotion.github.io/.
Authors: Minxue Tang, Yangyang Yu, Aolin Ding, Maziyar Baran Pouyan, Taha Belkhouja, Yujia Bao
Abstract: Recognizing implicit visual and textual patterns is essential in many real-world applications of modern AI. However, tackling long-tail pattern recognition tasks remains challenging for current pre-trained foundation models such as LLMs and VLMs. While finetuning pre-trained models can improve accuracy in recognizing implicit patterns, it is usually infeasible due to a lack of training data and high computational overhead. In this paper, we propose ADAMAB, an efficient embedding calibration framework for few-shot pattern recognition. To maximally reduce the computational costs, ADAMAB trains embedder-agnostic light-weight calibrators on top of fixed embedding models without accessing their parameters. To mitigate the need for large-scale training data, we introduce an adaptive data augmentation strategy based on the Multi-Armed Bandit (MAB) mechanism. With a modified upper confidence bound algorithm, ADAMAB diminishes the gradient shifting and offers theoretically guaranteed convergence in few-shot training. Our multi-modal experiments justify the superior performance of ADAMAB, with up to 40% accuracy improvement when training with less than 5 initial data samples of each class.
Authors: Jiabao Chen, Shan Xiong, Jialin Peng
Abstract: Domain adaptive segmentation (DAS) is a promising paradigm for delineating intracellular structures from various large-scale electron microscopy (EM) without incurring extensive annotated data in each domain. However, the prevalent unsupervised domain adaptation (UDA) strategies often demonstrate limited and biased performance, which hinders their practical applications. In this study, we explore sparse points and local human preferences as weak labels in the target domain, thereby presenting a more realistic yet annotation-efficient setting. Specifically, we develop Prefer-DAS, which pioneers sparse promptable learning and local preference alignment. The Prefer-DAS is a promptable multitask model that integrates self-training and prompt-guided contrastive learning. Unlike SAM-like methods, the Prefer-DAS allows for the use of full, partial, and even no point prompts during both training and inference stages and thus enables interactive segmentation. Instead of using image-level human preference alignment for segmentation, we introduce Local direct Preference Optimization (LPO) and sparse LPO (SLPO), plug-and-play solutions for alignment with spatially varying human feedback or sparse feedback. To address potential missing feedback, we also introduce Unsupervised Preference Optimization (UPO), which leverages self-learned preferences. As a result, the Prefer-DAS model can effectively perform both weakly-supervised and unsupervised DAS, depending on the availability of points and human preferences. Comprehensive experiments on four challenging DAS tasks demonstrate that our model outperforms SAM-like methods as well as unsupervised and weakly-supervised DAS methods in both automatic and interactive segmentation modes, highlighting strong generalizability and flexibility. Additionally, the performance of our model is very close to or even exceeds that of supervised models.
Authors: Yuxuan Yang, Zhonghao Yan, Yi Zhang, Bo Yun, Muxi Diao, Guowei Zhao, Kongming Liang, Wenbin Li, Zhanyu Ma
Abstract: Hepatocellular Carcinoma diagnosis relies heavily on the interpretation of gigapixel Whole Slide Images. However, current computational approaches are constrained by fixed-resolution processing mechanisms and inefficient feature aggregation, which inevitably lead to either severe information loss or high feature redundancy. To address these challenges, we propose Hepato-LLaVA, a specialized Multi-modal Large Language Model designed for fine-grained hepatocellular pathology analysis. We introduce a novel Sparse Topo-Pack Attention mechanism that explicitly models 2D tissue topology. This mechanism effectively aggregates local diagnostic evidence into semantic summary tokens while preserving global context. Furthermore, to overcome the lack of multi-scale data, we present HepatoPathoVQA, a clinically grounded dataset comprising 33K hierarchically structured question-answer pairs validated by expert pathologists. Our experiments demonstrate that Hepato-LLaVA achieves state-of-the-art performance on HCC diagnosis and captioning tasks, significantly outperforming existing methods. Our code and implementation details are available at https://pris-cv.github.io/Hepto-LLaVA/.
Authors: Zanxi Ruan, Songqun Gao, Qiuyu Kong, Yiming Wang, Marco Cristani
Abstract: Edge-based representations are fundamental cues for visual understanding, a principle rooted in early vision research and still central today. We extend this principle to vision-language alignment, showing that isolating and aligning structural cues across modalities can greatly benefit fine-tuning on long, detail-rich captions, with a specific focus on improving cross-modal retrieval. We introduce StructXLIP, a fine-tuning alignment paradigm that extracts edge maps (e.g., Canny), treating them as proxies for the visual structure of an image, and filters the corresponding captions to emphasize structural cues, making them "structure-centric". Fine-tuning augments the standard alignment loss with three structure-centric losses: (i) aligning edge maps with structural text, (ii) matching local edge regions to textual chunks, and (iii) connecting edge maps to color images to prevent representation drift. From a theoretical standpoint, while standard CLIP maximizes the mutual information between visual and textual embeddings, StructXLIP additionally maximizes the mutual information between multimodal structural representations. This auxiliary optimization is intrinsically harder, guiding the model toward more robust and semantically stable minima, enhancing vision-language alignment. Beyond outperforming current competitors on cross-modal retrieval in both general and specialized domains, our method serves as a general boosting recipe that can be integrated into future approaches in a plug-and-play manner. Code and pretrained models are publicly available at: https://github.com/intelligolabs/StructXLIP.
Authors: Chen Wang, Hao Tan, Wang Yifan, Zhiqin Chen, Yuheng Liu, Kalyan Sunkavalli, Sai Bi, Lingjie Liu, Yiwei Hu
Abstract: We propose tttLRM, a novel large 3D reconstruction model that leverages a Test-Time Training (TTT) layer to enable long-context, autoregressive 3D reconstruction with linear computational complexity, further scaling the model's capability. Our framework efficiently compresses multiple image observations into the fast weights of the TTT layer, forming an implicit 3D representation in the latent space that can be decoded into various explicit formats, such as Gaussian Splats (GS) for downstream applications. The online learning variant of our model supports progressive 3D reconstruction and refinement from streaming observations. We demonstrate that pretraining on novel view synthesis tasks effectively transfers to explicit 3D modeling, resulting in improved reconstruction quality and faster convergence. Extensive experiments show that our method achieves superior performance in feedforward 3D Gaussian reconstruction compared to state-of-the-art approaches on both objects and scenes.
Authors: Limai Jiang, Ruitao Xie, Bokai Yang, Huazhen Huang, Juan He, Yufu Huo, Zikai Wang, Yang Wei, Yunpeng Cai
Abstract: Medical image segmentation plays a vital role in clinical decision-making, enabling precise localization of lesions and guiding interventions. Despite significant advances in segmentation accuracy, the black-box nature of most deep models has raised growing concerns about their trustworthiness in high-stakes medical scenarios. Current explanation techniques have primarily focused on classification tasks, leaving the segmentation domain relatively underexplored. We introduced an explanation model for segmentation task which employs the causal inference framework and backpropagates the average treatment effect (ATE) into a quantification metric to determine the influence of input regions, as well as network components, on target segmentation areas. Through comparison with recent segmentation explainability techniques on two representative medical imaging datasets, we demonstrated that our approach provides more faithful explanations than existing approaches. Furthermore, we carried out a systematic causal analysis of multiple foundational segmentation models using our method, which reveals significant heterogeneity in perceptual strategies across different models, and even between different inputs for the same model. Suggesting the potential of our method to provide notable insights for optimizing segmentation models. Our code can be found at https://github.com/lcmmai/PdCR.
Authors: Chenyue Yu, Lingao Xiao, Jinhong Deng, Ivor W. Tsang, Yang He
Abstract: Large-scale image datasets are fundamental to deep learning, but their high storage demands pose challenges for deployment in resource-constrained environments. While existing approaches reduce dataset size by discarding samples, they often ignore the significant redundancy within each image -- particularly in the color space. To address this, we propose Dataset Color Quantization (DCQ), a unified framework that compresses visual datasets by reducing color-space redundancy while preserving information crucial for model training. DCQ achieves this by enforcing consistent palette representations across similar images, selectively retaining semantically important colors guided by model perception, and maintaining structural details necessary for effective feature learning. Extensive experiments across CIFAR-10, CIFAR-100, Tiny-ImageNet, and ImageNet-1K show that DCQ significantly improves training performance under aggressive compression, offering a scalable and robust solution for dataset-level storage reduction.
Authors: Zheng Gu, Min Lu, Zhida Sun, Dani Lischinski, Daniel Cohen-O, Hui Huang
Abstract: Disentangling visual layers in real-world images is a persistent challenge in vision and graphics, as such layers often involve non-linear and globally coupled interactions, including shading, reflection, and perspective distortion. In this work, we present an in-context image decomposition framework that leverages large diffusion foundation models for layered separation. We focus on the challenging case of logo-object decomposition, where the goal is to disentangle a logo from the surface on which it appears while faithfully preserving both layers. Our method fine-tunes a pretrained diffusion model via lightweight LoRA adaptation and introduces a cycle-consistent tuning strategy that jointly trains decomposition and composition models, enforcing reconstruction consistency between decomposed and recomposed images. This bidirectional supervision substantially enhances robustness in cases where the layers exhibit complex interactions. Furthermore, we introduce a progressive self-improving process, which iteratively augments the training set with high-quality model-generated examples to refine performance. Extensive experiments demonstrate that our approach achieves accurate and coherent decompositions and also generalizes effectively across other decomposition types, suggesting its potential as a unified framework for layered image decomposition.
Authors: Bowen Zheng, Yongli Xiang, Ziming Hong, Zerong Lin, Chaojian Yu, Tongliang Liu, Xinge You
Abstract: Image-to-Video (I2V) generation models, which condition video generation on reference images, have shown emerging visual instruction-following capability, allowing certain visual cues in reference images to act as implicit control signals for video generation. However, this capability also introduces a previously overlooked risk: adversaries may exploit visual instructions to inject malicious intent through the image modality. In this work, we uncover this risk by proposing Visual Instruction Injection (VII), a training-free and transferable jailbreaking framework that intentionally disguises the malicious intent of unsafe text prompts as benign visual instructions in the safe reference image. Specifically, VII coordinates a Malicious Intent Reprogramming module to distill malicious intent from unsafe text prompts while minimizing their static harmfulness, and a Visual Instruction Grounding module to ground the distilled intent onto a safe input image by rendering visual instructions that preserve semantic consistency with the original unsafe text prompt, thereby inducing harmful content during I2V generation. Empirically, our extensive experiments on four state-of-the-art commercial I2V models (Kling-v2.5-turbo, Gemini Veo-3.1, Seedance-1.5-pro, and PixVerse-V5) demonstrate that VII achieves Attack Success Rates of up to 83.5% while reducing Refusal Rates to near zero, significantly outperforming existing baselines.
Authors: Yifan Wang, Francesco Pittaluga, Zaid Tasneem, Chenyu You, Manmohan Chandraker, Ziyu Jiang
Abstract: Controllable driving scene generation is critical for realistic and scalable autonomous driving simulation, yet existing approaches struggle to jointly achieve photorealism and precise control. We introduce HorizonForge, a unified framework that reconstructs scenes as editable Gaussian Splats and Meshes, enabling fine-grained 3D manipulation and language-driven vehicle insertion. Edits are rendered through a noise-aware video diffusion process that enforces spatial and temporal consistency, producing diverse scene variations in a single feed-forward pass without per-trajectory optimization. To standardize evaluation, we further propose HorizonSuite, a comprehensive benchmark spanning ego- and agent-level editing tasks such as trajectory modifications and object manipulation. Extensive experiments show that Gaussian-Mesh representation delivers substantially higher fidelity than alternative 3D representations, and that temporal priors from video diffusion are essential for coherent synthesis. Combining these findings, HorizonForge establishes a simple yet powerful paradigm for photorealistic, controllable driving simulation, achieving an 83.4% user-preference gain and a 25.19% FID improvement over the second best state-of-the-art method. Project page: https://horizonforge.github.io/ .
Authors: Xihua Sheng, Lingyu Zhu, Tianyu Zhang, Dong Liu, Shiqi Wang, Jing Wang
Abstract: Diffusion-based generative image compression has demonstrated remarkable potential for achieving realistic reconstruction at ultra-low bitrates. The key to unlocking this potential lies in making the entire compression process content-adaptive, ensuring that the encoder's representation and the decoder's generative prior are dynamically aligned with the semantic and structural characteristics of the input image. However, existing methods suffer from three critical limitations that prevent effective content adaptation. First, isotropic quantization applies a uniform quantization step, failing to adapt to the spatially varying complexity of image content and creating a misalignment with the diffusion model's noise-dependent prior. Second, the information concentration bottleneck -- arising from the dimensional mismatch between the high-dimensional noisy latent and the diffusion decoder's fixed input -- prevents the model from adaptively preserving essential semantic information in the primary channels. Third, existing textual conditioning strategies either need significant textual bitrate overhead or rely on generic, content-agnostic textual prompts, thereby failing to provide adaptive semantic guidance efficiently. To overcome these limitations, we propose a content-adaptive diffusion-based image codec with three technical innovations: 1) an Uncertainty-Guided Adaptive Quantization method that learns spatial uncertainty maps to adaptively align quantization distortion with content characteristics; 2) an Auxiliary Decoder-Guided Information Concentration method that uses a lightweight auxiliary decoder to enforce content-aware information preservation in the primary latent channels; and 3) a Bitrate-Free Adaptive Textual Conditioning method that derives content-aware textual descriptions from the auxiliary reconstructed image, enabling semantic guidance without bitrate cost.
Authors: Shan Wang, Peixia Li, Chenchen Xu, Ziang Cheng, Jiayu Yang, Hongdong Li, Pulak Purkait
Abstract: We propose Light-Geometry Interaction (LGI) maps, a novel representation that encodes light-aware occlusion from monocular depth. Unlike ray tracing, which requires full 3D reconstruction, LGI captures essential light-shadow interactions reliably and accurately, computed from off-the-shelf 2.5D depth map predictions. LGI explicitly ties illumination direction to geometry, providing a physics-inspired prior that constrains generative models. Without such prior, these models often produce floating shadows, inconsistent illumination, and implausible shadow geometry. Building on this representation, we propose a unified pipeline for joint shadow generation and relighting - unlike prior methods that treat them as disjoint tasks - capturing the intrinsic coupling of illumination and shadowing essential for modeling indirect effects. By embedding LGI into a bridge-matching generative backbone, we reduce ambiguity and enforce physically consistent light-shadow reasoning. To enable effective training, we curated the first large-scale benchmark dataset for joint shadow and relighting, covering reflections, transparency, and complex interreflections. Experiments show significant gains in realism and consistency across synthetic and real images. LGI thus bridges geometry-inspired rendering with generative modeling, enabling efficient, physically consistent shadow generation and relighting.
Authors: Mingde Yao, Zhiyuan You, King-Man Tam, Menglu Wang, Tianfan Xue
Abstract: With the recent fast development of generative models, instruction-based image editing has shown great potential in generating high-quality images. However, the quality of editing highly depends on carefully designed instructions, placing the burden of task decomposition and sequencing entirely on the user. To achieve autonomous image editing, we present PhotoAgent, a system that advances image editing through explicit aesthetic planning. Specifically, PhotoAgent formulates autonomous image editing as a long-horizon decision-making problem. It reasons over user aesthetic intent, plans multi-step editing actions via tree search, and iteratively refines results through closed-loop execution with memory and visual feedback, without requiring step-by-step user prompts. To support reliable evaluation in real-world scenarios, we introduce UGC-Edit, an aesthetic evaluation benchmark consisting of 7,000 photos and a learned aesthetic reward model. We also construct a test set containing 1,017 photos to systematically assess autonomous photo editing performance. Extensive experiments demonstrate that PhotoAgent consistently improves both instruction adherence and visual quality compared with baseline methods. The project page is https://mdyao.github.io/PhotoAgent/.
Authors: Jiayu Chen, Ruoyu Lin, Zihao Zheng, Jingxin Li, Maoliang Li, Guojie Luo, Xiang Chen
Abstract: Visual Autoregressive(VAR) models enhance generation quality but face a critical efficiency bottleneck in later stages. In this paper, we present a novel optimization framework for VAR models that fundamentally differs from prior approaches such as FastVAR and SkipVAR. Instead of relying on heuristic skipping strategies, our method leverages attention entropy to characterize the semantic projections across different dimensions of the model architecture. This enables precise identification of parameter dynamics under varying token granularity levels, semantic scopes, and generation scales. Building on this analysis, we further uncover sparsity patterns along three critical dimensions-token, layer, and scale-and propose a set of fine-grained optimization strategies tailored to these patterns. Extensive evaluation demonstrates that our approach achieves aggressive acceleration of the generation process while significantly preserving semantic fidelity and fine details, outperforming traditional methods in both efficiency and quality. Experiments on Infinity-2B and Infinity-8B models demonstrate that ToProVAR achieves up to 3.4x acceleration with minimal quality loss, effectively mitigating the issues found in prior work. Our code will be made publicly available.
Authors: Tianyue Wang, Leigang Qu, Tianyu Yang, Xiangzhao Hao, Yifan Xu, Haiyun Guo, Jinqiao Wang
Abstract: Zero-Shot Composed Image Retrieval (ZS-CIR) aims to retrieve target images given a multimodal query (comprising a reference image and a modification text), without training on annotated triplets. Existing methods typically convert the multimodal query into a single modality-either as an edited caption for Text-to-Image retrieval (T2I) or as an edited image for Image-to-Image retrieval (I2I). However, each paradigm has inherent limitations: T2I often loses fine-grained visual details, while I2I struggles with complex semantic modifications. To effectively leverage their complementary strengths under diverse query intents, we propose WISER, a training-free framework that unifies T2I and I2I via a "retrieve-verify-refine" pipeline, explicitly modeling intent awareness and uncertainty awareness. Specifically, WISER first performs Wider Search by generating both edited captions and images for parallel retrieval to broaden the candidate pool. Then, it conducts Adaptive Fusion with a verifier to assess retrieval confidence, triggering refinement for uncertain retrievals, and dynamically fusing the dual-path for reliable ones. For uncertain retrievals, WISER generates refinement suggestions through structured self-reflection to guide the next retrieval round toward Deeper Thinking. Extensive experiments demonstrate that WISER significantly outperforms previous methods across multiple benchmarks, achieving relative improvements of 45% on CIRCO (mAP@5) and 57% on CIRR (Recall@1) over existing training-free methods. Notably, it even surpasses many training-dependent methods, highlighting its superiority and generalization under diverse scenarios. Code will be released at https://github.com/Physicsmile/WISER.
Authors: Zhaochen Su, Jincheng Gao, Hangyu Guo, Zhenhua Liu, Lueyang Zhang, Xinyu Geng, Shijue Huang, Peng Xia, Guanyu Jiang, Cheng Wang, Yue Zhang, Yi R. Fung, Junxian He
Abstract: Real-world multimodal agents solve multi-step workflows grounded in visual evidence. For example, an agent can troubleshoot a device by linking a wiring photo to a schematic and validating the fix with online documentation, or plan a trip by interpreting a transit map and checking schedules under routing constraints. However, existing multimodal benchmarks mainly evaluate single-turn visual reasoning or specific tool skills, and they do not fully capture the realism, visual subtlety, and long-horizon tool use that practical agents require. We introduce AgentVista, a benchmark for generalist multimodal agents that spans 25 sub-domains across 7 categories, pairing realistic and detail-rich visual scenarios with natural hybrid tool use. Tasks require long-horizon tool interactions across modalities, including web search, image search, page navigation, and code-based operations for both image processing and general programming. Comprehensive evaluation of state-of-the-art models exposes significant gaps in their ability to carry out long-horizon multimodal tool use. Even the best model in our evaluation, Gemini-3-Pro with tools, achieves only 27.3% overall accuracy, and hard instances can require more than 25 tool-calling turns. We expect AgentVista to accelerate the development of more capable and reliable multimodal agents for realistic and ultra-challenging problem solving.
Authors: Ziqi Gao, Jieyu Zhang, Wisdom Oluchi Ikezogwo, Jae Sung Park, Tario G. You, Daniel Ogbu, Chenhao Zheng, Weikai Huang, Yinuo Yang, Winson Han, Quan Kong, Rajat Saini, Ranjay Krishna
Abstract: We introduce Synthetic Visual Genome 2 (SVG2), a large-scale panoptic video scene graph dataset. SVG2 contains over 636K videos with 6.6M objects, 52.0M attributes, and 6.7M relations, providing an order-of-magnitude increase in scale and diversity over prior spatio-temporal scene graph datasets. To create SVG2, we design a fully automated pipeline that combines multi-scale panoptic segmentation, online-offline trajectory tracking with automatic new-object discovery, per-trajectory semantic parsing, and GPT-5-based spatio-temporal relation inference. Building on this resource, we train TRaSER, a video scene graph generation model. TRaSER augments VLMs with a trajectory-aligned token arrangement mechanism and new modules: an object-trajectory resampler and a temporal-window resampler to convert raw videos and panoptic trajectories into compact spatio-temporal scene graphs in a single forward pass. The temporal-window resampler binds visual tokens to short trajectory segments to preserve local motion and temporal semantics, while the object-trajectory resampler aggregates entire trajectories to maintain global context for objects. On the PVSG, VIPSeg, VidOR and SVG2 test datasets, TRaSER improves relation detection by +15 to 20%, object prediction by +30 to 40% over the strongest open-source baselines and by +13% over GPT-5, and attribute prediction by +15%. When TRaSER's generated scene graphs are sent to a VLM for video question answering, it delivers a +1.5 to 4.6% absolute accuracy gain over using video only or video augmented with Qwen2.5-VL's generated scene graphs, demonstrating the utility of explicit spatio-temporal scene graphs as an intermediate representation.
Authors: Hiroshi Sasaki
Abstract: Recent multimodal models such as Contrastive Language-Image Pre-training (CLIP) have shown remarkable ability to align visual and linguistic representations. However, domains where small visual differences carry large semantic significance, such as diagram understanding, remain challenging due to the models' limited sensitivity to fine-grained structural variations. We propose a new training paradigm designed to enhance diagram comprehension in vision-language models. Our approach introduces pseudo contrastive samples generated by a diagram renderer that creates synthetic diagrams using randomly picked text elements. These samples highlight structural differences in diagrammatic imagery without requiring any modification or editing of the original data. By incorporating these pseudo contrastive samples into the training objective, the model learns to capture more precise and semantically consistent diagram structures. Empirical evaluations on a benchmark dataset of flowcharts demonstrate substantial improvements over standard CLIP and hard-negative CLIP training in both image-text matching and visual question answering tasks. The results underscore the value of domain-specific training strategies and contribute to advancing diagrammatic understanding within the broader context of vision-language learning.
Authors: Jiahui Zhan, Xianbing Sun, Xiangnan Zhu, Yikun Ji, Ruitong Liu, Liqing Zhang, Jianfu Zhang
Abstract: Object swapping aims to replace a source object in a scene with a reference object while preserving object fidelity, scene fidelity, and object-scene harmony. Existing methods either require per-object finetuning and slow inference or rely on extra paired data that mostly depict the same object across contexts, forcing models to rely on background cues rather than learning cross-object alignment. We propose SourceSwap, a self-supervised and source-aware framework that learns cross-object alignment. Our key insight is to synthesize high-quality pseudo pairs from any image via a frequency-separated perturbation in the initial-noise space, which alters appearance while preserving pose, coarse shape, and scene layout, requiring no videos, multi-view data, or additional images. We then train a dual U-Net with full-source conditioning and a noise-free reference encoder, enabling direct inter-object alignment, zero-shot inference without per-object finetuning, and lightweight iterative refinement. We further introduce SourceBench, a high-quality benchmark with higher resolution, more categories, and richer interactions. Experiments demonstrate that SourceSwap achieves superior fidelity, stronger scene preservation, and more natural harmony, and it transfers well to edits such as subject-driven refinement and face swapping.
Authors: Bowen Yang, Zishuo Li, Yang Sun, Changtao Miao, Yifan Yang, Man Luo, Xiaotong Yan, Feng Jiang, Jinchuan Shi, Yankai Fu, Ning Chen, Junkai Zhao, Pengwei Wang, Guocai Yao, Shanghang Zhang, Hao Chen, Zhe Li, Kai Zhu
Abstract: Embodied foundation models require large-scale, high-quality real-world interaction data for pre-training and scaling. However, existing data collection methods suffer from high infrastructure costs, complex hardware dependencies, and limited interaction scope, making scalable expansion challenging. In fact, humans themselves are ideal physically embodied agents. Therefore, obtaining egocentric real-world interaction data from globally distributed "human agents" offers advantages of low cost and sustainability. To this end, we propose the Always-on Egocentric (AoE) data collection system, which aims to simplify hardware dependencies by leveraging humans themselves and their smartphones, enabling low-cost, highly efficient, and scene-agnostic real-world interaction data collection to address the challenge of data scarcity. Specifically, we first employ an ergonomic neck-mounted smartphone holder to enable low-barrier, large-scale egocentric data collection through a cloud-edge collaborative architecture. Second, we develop a cross-platform mobile APP that leverages on-device compute for real-time processing, while the cloud hosts automated labeling and filtering pipelines that transform raw videos into high-quality training data. Finally, the AoE system supports distributed Ego video data collection by anyone, anytime, and anywhere. We evaluate AoE on data preprocessing quality and downstream tasks, demonstrating that high-quality egocentric data significantly boosts real-world generalization.
Authors: Salome Kazeminia, Carsten Marr, Bastian Rieck
Abstract: Multiple instance learning (MIL) is a framework for weakly supervised classification, where labels are assigned to sets of instances, i.e., bags, rather than to individual data points. This paradigm has proven effective in tasks where fine-grained annotations are unavailable or costly to obtain. However, the effectiveness of MIL drops sharply when training data are scarce, such as for rare disease classification. To address this challenge, we propose incorporating topological inductive biases into the data representation space within the MIL framework. This bias introduces a topology-preserving constraint that encourages the instance encoder to maintain the topological structure of the instance distribution within each bag when mapping them to MIL latent space. As a result, our Topology Guided MIL (TG-MIL) method enhances the performance and generalizability of MIL classifiers across different aggregation functions, especially under scarce-data regimes. Our evaluations show average performance improvements of 15.3% for synthetic MIL datasets, 2.8% for MIL benchmarks, and 5.5% for rare anemia classification compared to current state-of-the-art MIL models, where only 17-120 samples per class are available. We make our code publicly available.
Authors: Yuhang Liu, Zhen Zhang, Dong Gong, Erdun Gao, Biwei Huang, Mingming Gong, Anton van den Hengel, Kun Zhang, Javen Qinfeng Shi
Abstract: Directed Acyclic Graphs (DAGs) are a standard tool in causal modeling, but their suitability for capturing the complexity of large-scale multimodal data is questionable. In practice, real-world multimodal datasets are often collected from heterogeneous generative processes that do not conform to a single DAG. Instead, they may involve multiple, and even opposing, DAG structures with inverse causal directions. To address this gap, in this work, we first propose a novel latent partial causal model tailored for multimodal data representation learning, featuring two latent coupled variables parts connected by an undirected edge, to represent the transfer of knowledge across modalities. Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by MultiModal Contrastive Learning (MMCL) correspond to the latent coupled variables up to a trivial transformation. This result deepens our understanding of the why MMCL works, highlights its potential for representation disentanglement, and expands the utility of pre-trained models like CLIP. Synthetic experiments confirm the robustness of our findings, even when the assumptions are partially violated. Most importantly, experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets. Together, these contributions push the boundaries of MMCL, both in theory and in practical applications.
Authors: Amir Mohammad Naderi, Jennifer G. Casey, Mao-Hsiang Huang, Rachelle Victorio, David Y. Chiang, Calum MacRae, Hung Cao, Vandana A. Gupta
Abstract: Quantifying cardiovascular parameters like ejection fraction in zebrafish as a host of biological investigations has been extensively studied. Since current manual monitoring techniques are time-consuming and fallible, several image processing frameworks have been proposed to automate the process. Most of these works rely on supervised deep-learning architectures. However, supervised methods tend to be overfitted on their training dataset. This means that applying the same framework to new data with different imaging setups and mutant types can severely decrease performance. We have developed a Zebrafish Automatic Cardiovascular Assessment Framework (ZACAF) to quantify the cardiac function in zebrafish. In this work, we further applied data augmentation, Transfer Learning (TL), and Test Time Augmentation (TTA) to ZACAF to improve the performance for the quantification of cardiovascular function quantification in zebrafish. This strategy can be integrated with the available frameworks to aid other researchers. We demonstrate that using TL, even with a constrained dataset, the model can be refined to accommodate a novel microscope setup, encompassing diverse mutant types and accommodating various video recording protocols. Additionally, as users engage in successive rounds of TL, the model is anticipated to undergo substantial enhancements in both generalizability and accuracy. Finally, we applied this approach to assess the cardiovascular function in nrap mutant zebrafish, a model of cardiomyopathy.
Authors: Dang Nguyen, Paymon Haddad, Eric Gan, Baharan Mirzasoleiman
Abstract: Can we modify the training data distribution to encourage the underlying optimization method toward finding solutions with superior generalization performance on in-distribution data? In this work, we approach this question for the first time by comparing the inductive bias of gradient descent (GD) with that of sharpness-aware minimization (SAM). By studying a two-layer CNN, we rigorously prove that SAM learns different features more uniformly, particularly in early epochs. That is, SAM is less susceptible to simplicity bias compared to GD. We also show that examples containing features that are learned early are separable from the rest based on the model's output. Based on this observation, we propose a method that (i) clusters examples based on the network output early in training, (ii) identifies a cluster of examples with similar network output, and (iii) upsamples the rest of examples only once to alleviate the simplicity bias. We show empirically that USEFUL effectively improves the generalization performance on the original data distribution when training with various gradient methods, including (S)GD and SAM. Notably, we demonstrate that our method can be combined with SAM variants and existing data augmentation strategies to achieve, to the best of our knowledge, state-of-the-art performance for training ResNet18 on CIFAR10, STL10, CINIC10, Tiny-ImageNet; ResNet34 on CIFAR100; and VGG19 and DenseNet121 on CIFAR10. Our code is available at https://github.com/BigML-CS-UCLA/TADA.
Authors: Rang Meng, Xingyu Zhang, Yuming Li, Chenguang Ma
Abstract: Recent work on human animation usually involves audio, pose, or movement maps conditions, thereby achieves vivid animation quality. However, these methods often face practical challenges due to extra control conditions, cumbersome condition injection modules, or limitation to head region driving. Hence, we ask if it is possible to achieve striking half-body human animation while simplifying unnecessary conditions. To this end, we propose a half-body human animation method, dubbed EchoMimicV2, that leverages a novel Audio-Pose Dynamic Harmonization strategy, including Pose Sampling and Audio Diffusion, to enhance half-body details, facial and gestural expressiveness, and meanwhile reduce conditions redundancy. To compensate for the scarcity of half-body data, we utilize Head Partial Attention to seamlessly accommodate headshot data into our training framework, which can be omitted during inference, providing a free lunch for animation. Furthermore, we design the Phase-specific Denoising Loss to guide motion, detail, and low-level quality for animation in specific phases, respectively. Besides, we also present a novel benchmark for evaluating the effectiveness of half-body human animation. Extensive experiments and analyses demonstrate that EchoMimicV2 surpasses existing methods in both quantitative and qualitative evaluations.
Authors: Shadab Ahamed, Eldad Haber
Abstract: Inverse problems, which involve estimating parameters from incomplete or noisy observations, arise in various fields such as medical imaging, geophysics, and signal processing. These problems are often ill-posed, requiring regularization techniques to stabilize the solution. In this work, we employ Flow Matching (FM), a generative framework that integrates a deterministic processes to map a simple reference distribution, such as a Gaussian, to the target distribution. Our method DAWN-FM: Data-AWare and Noise-Informed Flow Matching incorporates data and noise embedding, allowing the model to access representations about the measured data explicitly and also account for noise in the observations, making it particularly robust in scenarios where data is noisy or incomplete. By learning a time-dependent velocity field, FM not only provides accurate solutions but also enables uncertainty quantification by generating multiple plausible outcomes. Unlike pretrained diffusion models, which may struggle in highly ill-posed settings, our approach is trained specifically for each inverse problem and adapts to varying noise levels. We validate the effectiveness and robustness of our method through extensive numerical experiments on tasks such as image deblurring and tomography. The code is available at: https://github.com/ahxmeds/DAWN-FM.git.
Authors: Ismail Khalfaoui-Hassani, Stefan Kesselheim
Abstract: Which functions can be used as activations in deep neural networks? This article explores families of functions based on orthonormal bases, including the Hermite polynomial basis and the Fourier trigonometric basis, as well as a basis resulting from the tropicalization of a polynomial basis. Our study shows that, through simple variance-preserving initialization and without additional clamping mechanisms, these activations can successfully be used to train deep models, such as GPT-2 for next-token prediction on OpenWebText and ConvNeXt for image classification on ImageNet. Our work addresses the issue of exploding and vanishing activations and gradients, particularly prevalent with polynomial activations, and opens the door for improving the efficiency of large-scale learning tasks. Furthermore, our approach provides insight into the structure of neural networks, revealing that networks with polynomial activations can be interpreted as multivariate polynomial mappings. Finally, using Hermite interpolation, we show that our activations can closely approximate classical ones in pre-trained models by matching both the function and its derivative, making them especially useful for fine-tuning tasks. These activations are available in the torchortho library via: https://github.com/K-H-Ismail/torchortho.
Authors: Chayanin Tangwiriyasakul, Pedro Borges, Guilherme Pombo, Stefano Moriconi, Michael S. Elmalem, Paul Wright, Yee-Haur Mah, Jane Rondina, Sebastien Ourselin, Parashkev Nachev, M. Jorge Cardoso
Abstract: Focal deficits in ischaemic stroke result from impaired perfusion downstream of a critical vascular occlusion. While parenchymal lesions are traditionally used to predict clinical deficits, the underlying pattern of disrupted perfusion provides information upstream of the lesion, potentially yielding earlier predictive and localizing signals. Such perfusion maps can be derived from routine CT angiography (CTA) widely deployed in clinical practice. Analysing computed perfusion maps from 1,393 CTA-imaged-patients with acute ischaemic stroke, we use deep generative inference to localise neural substrates of NIHSS sub-scores. We show that our approach replicates known lesion-deficit relations without knowledge of the lesion itself and reveals novel neural dependents. The high achieved anatomical fidelity suggests acute CTA-derived computed perfusion maps may be of substantial clinical-and-scientific value in rich phenotyping of acute stroke. Using only hyperacute imaging, deep generative inference could power highly expressive models of functional anatomical relations in ischaemic stroke within the pre-interventional window.
Authors: G\"okhan \"Ozbulak, Oscar Jimenez-del-Toro, Ma\'ira Fatoretto, Lilian Berton, Andr\'e Anjos
Abstract: The evaluation of fairness models in Machine Learning involves complex challenges, such as defining appropriate metrics, balancing trade-offs between utility and fairness, and there are still gaps in this stage. This work presents a novel multi-objective evaluation framework that enables the analysis of utility-fairness trade-offs in Machine Learning systems. The framework was developed using criteria from Multi-Objective Optimization that collect comprehensive information regarding this complex evaluation task. The assessment of multiple Machine Learning systems is summarized, both quantitatively and qualitatively, in a straightforward manner through a radar chart and a measurement table encompassing various aspects such as convergence, system capacity, and diversity. The framework's compact representation of performance facilitates the comparative analysis of different Machine Learning strategies for decision-makers, in real-world applications, with single or multiple fairness requirements. In particular, this study focuses on the medical imaging domain, where fairness considerations are crucial due to the potential impact of biased diagnostic systems on patient outcomes. The proposed framework enables a systematic evaluation of multiple fairness constraints helping to identify and mitigate disparities among demographic groups while maintaining diagnostic performance. The framework is model-agnostic and flexible to be adapted to any kind of Machine Learning systems, that is, black- or white-box, any kind and quantity of evaluation metrics, including multidimensional fairness criteria. The functionality and effectiveness of the proposed framework is shown with different simulations, and an empirical study conducted on three real-world medical imaging datasets with various Machine Learning systems. Our evaluation framework is publicly available at https://pypi.org/project/fairical.
Authors: Chanhyuk Lee, Jiho Choi, Chanryeol Lee, Donggyun Kim, Seunghoon Hong
Abstract: Model merging has emerged as a promising approach for unifying independently fine-tuned models into an integrated framework, significantly enhancing computational efficiency in multi-task learning. Recently, several SVD-based techniques have been introduced to exploit low-rank structures for enhanced merging, but their reliance on such manually designed rank selection often leads to cross-task interference and suboptimal performance. In this paper, we propose AdaRank, a novel model merging framework that adaptively selects the most beneficial singular directions of task vectors to merge multiple models. We empirically show that the dominant singular components of task vectors can cause critical interference with other tasks, and that naive truncation across tasks and layers degrades performance. In contrast, AdaRank dynamically prunes the singular components that cause interference and offers an optimal amount of information to each task vector by learning to prune ranks during test-time via entropy minimization. Our analysis demonstrates that such method mitigates detrimental overlaps among tasks, while empirical results show that AdaRank consistently achieves state-of-the-art performance with various backbones and number of tasks, reducing the performance gap between fine-tuned models to nearly 1%.
Authors: Jiude Wei, Yuxuan Li, Cewu Lu, Jianhua Sun
Abstract: We humans rely on a wide range of commonsense knowledge to interact with an extensive number and categories of objects in the physical world. Likewise, such commonsense knowledge is also crucial for robots to successfully develop generalized object manipulation skills. While recent advancements in Multi-modal Large Language Models (MLLMs) have showcased their impressive capabilities in acquiring commonsense knowledge and conducting commonsense reasoning, effectively grounding this semantic-level knowledge produced by MLLMs to the physical world to thoroughly guide robots in generalized articulated object manipulation remains a challenge that has not been sufficiently addressed. To this end, we introduce analytic concepts, procedurally defined upon mathematical symbolism that can be directly computed and simulated by machines. By leveraging the analytic concepts as a bridge between the semantic-level knowledge inferred by MLLMs and the physical world where real robots operate, we can figure out the knowledge of object structure and functionality with physics-informed representations, and then use the physically grounded knowledge to instruct robot control policies for generalized and accurate articulated object manipulation. Extensive experiments in both real world and simulation demonstrate the superiority of our approach.
Authors: Philip Naumann, Jacob Kauffmann, Gr\'egoire Montavon
Abstract: Wasserstein distances provide a powerful framework for comparing data distributions. They can be used to analyze processes over time or to detect inhomogeneities within data. However, simply calculating the Wasserstein distance or analyzing the corresponding transport plan (or coupling) may not be sufficient for understanding what factors contribute to a high or low Wasserstein distance. In this work, we propose a novel solution based on Explainable AI that allows us to efficiently and accurately attribute Wasserstein distances to various data components, including data subgroups, input features, or interpretable subspaces. Our method achieves high accuracy across diverse datasets and Wasserstein distance specifications, and its practical utility is demonstrated in three use cases.
Authors: Alonso Urbano, David W. Romero, Max Zimmer, Sebastian Pokutta
Abstract: Real world data often exhibits unknown, instance-specific symmetries that rarely exactly match a transformation group $G$ fixed a priori. Class-pose decompositions aim to create disentangled representations by factoring inputs into invariant features and a pose $g\in G$ defined relative to a training-dependent, arbitrary canonical representation. We introduce RECON, a class-pose agnostic canonical orientation normalization that corrects arbitrary canonicals via a simple right translation, yielding natural, data-aligned canonicalizations. This enables (i) unsupervised discovery of instance-specific pose distributions, (ii) detection of out-of-distribution poses and (iii) a plug-and-play test-time canonicalization layer. This layer can be attached on top of any pre-trained model to infuse group invariance, improving its performance without retraining. We validate on images and molecular ensembles, demonstrating accurate symmetry discovery, and matching or outperforming other canonicalizations in downstream classification.
Authors: Soichiro Kumano, Hiroshi Kera, Toshihiko Yamasaki
Abstract: Adversarial training is one of the most effective defenses against adversarial attacks, but it incurs a high computational cost. In this study, we present the first theoretical analysis suggesting that adversarially pretrained transformers can serve as universally robust foundation models -- models that can adapt robustly to diverse downstream tasks with only lightweight tuning. Specifically, we demonstrate that single-layer linear transformers, after adversarial pretraining across a variety of classification tasks, can generalize robustly to unseen classification tasks through in-context learning from clean demonstrations (i.e., without requiring additional adversarial training or examples). This universal robustness stems from the model's ability to adaptively focus on robust features within given tasks. We also identify two open challenges for attaining robustness: the accuracy-robustness trade-off and sample-hungry training. This study initiates the discussion on the utility of universally robust foundation models. While their training is expensive, the investment would prove worthwhile as downstream tasks can obtain adversarial robustness for free. The code is available at https://github.com/s-kumano/universally-robust-in-context-learner.
URLs: https://github.com/s-kumano/universally-robust-in-context-learner.
Authors: Mariia Seleznova, Hung-Hsu Chou, Claudio Mayrink Verdun, Gitta Kutyniok
Abstract: We introduce GradPCA, an Out-of-Distribution (OOD) detection method that exploits the low-rank structure of neural network gradients induced by Neural Tangent Kernel (NTK) alignment. GradPCA applies Principal Component Analysis (PCA) to gradient class-means, achieving more consistent performance than existing methods across standard image classification benchmarks. We provide a theoretical perspective on spectral OOD detection in neural networks to support GradPCA, highlighting feature-space properties that enable effective detection and naturally emerge from NTK alignment. Our analysis further reveals that feature quality -- particularly the use of pretrained versus non-pretrained representations -- plays a crucial role in determining which detectors will succeed. Extensive experiments validate the strong performance of GradPCA, and our theoretical framework offers guidance for designing more principled spectral OOD detectors.
Authors: Kejing Lu, Chuan Xiao, Yoshiharu Ishikawa
Abstract: In this paper, we study the angle testing problem in the context of similarity search in high-dimensional Euclidean spaces and propose two projection-based probabilistic kernel functions, one designed for angle comparison and the other for angle thresholding. Unlike existing approaches that rely on random projection vectors drawn from Gaussian distributions, our approach leverages reference angles and adopts a deterministic structure for the projection vectors. Notably, our kernel functions do not require asymptotic assumptions, such as the number of projection vectors tending to infinity, and can be theoretically and experimentally shown to outperform Gaussian-distribution-based kernel functions. We apply the proposed kernel function to Approximate Nearest Neighbor Search (ANNS) and demonstrate that our approach achieves a 2.5x--3x higher query-per-second (QPS) throughput compared to the widely-used graph-based search algorithm HNSW.
Authors: Akash Gupta, Amos Storkey, Mirella Lapata
Abstract: Large Multimodal Models (LMMs) often rely on in-context learning (ICL) to perform new visual question answering (VQA) tasks with minimal supervision. However, ICL performance, especially in smaller LMMs, does not always improve monotonically when increasing the number of examples. We hypothesize that this happens because the LMM is overwhelmed by extraneous information in the image embeddings that is irrelevant to the downstream task. To address this, we propose a meta-learning approach that induces few-shot capabilities in LMMs through a fixed set of soft prompts distilled from task-relevant visual features, which are adapted at test time using a small number of examples. We facilitate this distillation through an attention-mapper module that can be easily integrated with any LMM architecture and is jointly learned with soft prompts. Evaluation on the VL-ICL Bench shows that our method successfully achieves task adaptation in low-data regimes with just a few gradient steps, outperforming ICL by 21.2%. Comparisons with parameter-efficient finetuning methods demonstrate that meta-learning further enhances this adaptation by 7.7% for various VQA tasks.
Authors: Hongyu Wang, Chuyan Xiong, Ruiping Wang, Xilin Chen
Abstract: Deploying powerful Vision-Language-Action (VLA) models on edge devices is limited by their massive size. In this paper, we take a deployment-oriented view of VLA training: we target efficiency through model design and optimization, rather than relying solely on post-hoc compression. Thus, we propose BitVLA, a fully native 1-bit VLA model for robotic manipulation, where every parameters is ternary, i.e., {-1,0,1}. BitVLA is built on the publicly available 1-bit LLM BitNet b1.58 2B4T, and is trained as a vision-language-action policy that inherits the compactness of 1-bit pretraining while retaining strong task performance. To further reduce the memory footprint of the vision backbone, we introduce Quantize-then-Distill, a post-training quantization-aware training strategy that compresses a full-precision vision encoder to 1.58-bit weights, while a full-precision teacher guides representation alignment during training. Across simulation benchmarks and real-world tasks, BitVLA matches the performance of the full-precision OpenVLA-OFT baseline, while reducing model memory by 11.0x and end-to-end latency by 4.4x. These results suggest a practical path toward training-time efficiency-accuracy co-design for embodied policies, enabling competitive manipulation capability on memory-constrained edge robotic platforms. We release the code in https://github.com/ustcwhy/BitVLA, model weights in https://huggingface.co/lxsy/bitvla-bf16.
URLs: https://github.com/ustcwhy/BitVLA,, https://huggingface.co/lxsy/bitvla-bf16.
Authors: Georgii Bychkov, Khaled Abud, Egor Kovalev, Alexander Gushchin, Sergey Lavrushkin, Dmitriy Vatolin, Anastasia Antsiferova
Abstract: Neural image compression (NIC) is increasingly used in computer vision pipelines, as learning-based models are able to surpass traditional algorithms in compression efficiency. However, learned codecs can be unstable and vulnerable to adversarial attacks: small perturbations may cause severe reconstruction artifacts or indirectly break downstream models. Despite these risks, most NIC benchmarks only emphasize rate-distortion (RD) performance, focusing on model efficiency in safe, non-adversarial scenarios, while NIC robustness studies cover only specific codecs and attacks. To fill this gap, we introduce \textbf{NIC-RobustBench}, an open-source benchmark and evaluation framework for adversarial robustness of NIC methods. The benchmark integrates 8 attacks, 9 defense strategies, standard RD metrics, a large and extensible set of codecs, and tools for assessing both the robustness of the compression model and impact on downstream tasks. Using NIC-RobustBench, we provide a broad empirical study of modern NICs and defenses in adversarial scenarios, highlighting failure modes, least and most resilient architectures, and other insights into NIC robustness. Our code is available online at https://github.com/msu-video-group/NIC-RobustBench.
Authors: Qinzhe Wang, Zixuan Chen, Keke Huang, Xiu Su, Chunhua Yang, Chang Xu
Abstract: Few-Shot Class Incremental Learning (FSCIL) is crucial for adapting to the complex open-world environments. Contemporary prospective learning-based space construction methods struggle to balance old and new knowledge, as prototype bias and rigid structures limit the expressive capacity of the embedding space. Different from these strategies, we rethink the optimization dilemma from the perspective of feature-structure dual consistency, and propose a Consistency-driven Calibration and Matching (ConCM) framework that systematically mitigates the knowledge conflict inherent in FSCIL. Specifically, inspired by hippocampal associative memory, we design a memory-aware prototype calibration that extracts generalized semantic attributes from base classes and reintegrates them into novel classes to enhance the conceptual center consistency of features. Further, to consolidate memory associations, we propose dynamic structure matching, which adaptively aligns the calibrated features to a session-specific optimal manifold space, ensuring cross-session structure consistency. This process requires no class-number priors and is theoretically guaranteed to achieve geometric optimality and maximum matching. On large-scale FSCIL benchmarks including mini-ImageNet, CIFAR100 and CUB200, ConCM achieves state-of-the-art performance, with harmonic accuracy gains of up to 3.41% in incremental sessions. Code is available at: https://github.com/wire-wqz/ConCM
Authors: Shai Yehezkel, Omer Dahary, Andrey Voynov, Daniel Cohen-Or
Abstract: Denoising diffusion models excel at generating high-quality images conditioned on text prompts, yet their effectiveness heavily relies on careful guidance during the sampling process. Classifier-Free Guidance (CFG) provides a widely used mechanism for steering generation by setting the guidance scale, which balances image quality and prompt alignment. However, the choice of the guidance scale has a critical impact on the convergence toward a visually appealing and prompt-adherent image. In this work, we propose an annealing guidance scheduler which dynamically adjusts the guidance scale over time based on the conditional noisy signal. By learning a scheduling policy, our method addresses the temperamental behavior of CFG. Empirical results demonstrate that our guidance scheduler significantly enhances image quality and alignment with the text prompt, advancing the performance of text-to-image generation. Notably, our novel scheduler requires no additional activations or memory consumption, and can seamlessly replace the common classifier-free guidance, offering an improved trade-off between prompt alignment and quality.
Authors: Shrunal Pothagoni, Benjamin Schweinhart
Abstract: Convolutional neural networks (CNNs) are a standard tool for computer vision tasks such as image classification. However, typical model architectures may result in the loss of topological information. In specific domains such as histopathology, topology is an important descriptor that can be used to distinguish between disease-indicating tissue by analyzing the shape characteristics of cells. Current literature suggests that reintroducing topological information using persistent homology can improve medical diagnostics; however, previous methods utilize global topological summaries which do not contain information about the locality of topological features. To address this gap, we present a novel method that generates local persistent homology-based data using a modified version of the convolution operator called Persistent Homology Convolutions. This method captures information about the locality and translation invariance of topological features. We perform a comparative study using various representations of histopathology slides and find that models trained with persistent homology convolutions outperform conventionally trained models and are less sensitive to hyperparameters. These results indicate that persistent homology convolutions extract meaningful geometric information from the histopathology slides.
Authors: Z. Ahmed, E. Bates, P. Francesch Huc, S. Y. W. Low, A. Golan, T. Bell, A. Rizza, S. D'Amico
Abstract: Spacecraft Rendezvous, Proximity Operations (RPO), and Formation Flying (FF) rely on safety-critical guidance, navigation and control (GNC) that must satisfy stringent performance and robustness requirements. However, verifying GNC performance is challenging due to the complexity and inaccessibility of the space environment, necessitating a verification and validation (V\&V) process that bridges simulation and real-world behavior. This paper contributes a unified, closed-loop, end-to-end digital and robotic twinning framework that enables software- and hardware-in-the-loop testing of spacecraft GNC systems. The framework is designed for modularity and flexibility, supporting interchangeable sensing modalities, control algorithms, and operational regimes. The digital twin includes an event-driven faster-than-real-time simulation environment to support rapid prototyping. The architecture is augmented with hardware-based robotic testbeds from Stanford's Space Rendezvous Laboratory (SLAB): the GNSS and Radiofrequency Autonomous Navigation Testbed for Distributed Space Systems (GRAND) to validate RF-based navigation techniques, and the Testbed for Rendezvous and Optical Navigation (TRON) and Optical Stimulator (OS) to validate vision-based methods. The test article for this work is an integrated multi-modal GNC software stack developed at SLAB. This paper introduces the hybrid twinning framework, summarizes calibration and error characterization of the robotic testbeds, and evaluates GNC performance across multiple operational modes in a full-range RPO scenario in LEO. The results demonstrate consistency between software- and hardware-in-the-loop tests with clear explainability for deviations in performance, thus validating the hybrid twinning pipeline as a reliable framework for realistic assessment and verification of GNC systems.
Authors: Yupei Zhang, Xiaofei Wang, Anran Liu, Lequan Yu, Chao Li
Abstract: Histopathology remains the gold standard for cancer diagnosis and prognosis. With the advent of transcriptome profiling, multi-modal learning combining transcriptomics with histology offers more comprehensive information. However, existing multi-modal approaches are challenged by intrinsic multi-modal heterogeneity, insufficient multi-scale integration, and reliance on paired data, restricting clinical applicability. To address these challenges, we propose a disentangled multi-modal framework with four contributions: 1) To mitigate multi-modal heterogeneity, we decompose WSIs and transcriptomes into tumor and microenvironment subspaces using a disentangled multi-modal fusion module, and introduce a confidence-guided gradient coordination strategy to balance subspace optimization. 2) To enhance multi-scale integration, we propose an inter-magnification gene-expression consistency strategy that aligns transcriptomic signals across WSI magnifications. 3) To reduce dependency on paired data, we propose a subspace knowledge distillation strategy enabling transcriptome-agnostic inference through a WSI-only student model. 4) To improve inference efficiency, we propose an informative token aggregation module that suppresses WSI redundancy while preserving subspace semantics. Extensive experiments on cancer diagnosis, prognosis, and survival prediction demonstrate our superiority over state-of-the-art methods across multiple settings. Code is available at https://github.com/helenypzhang/Disentangled-Multimodal-Learning.
URLs: https://github.com/helenypzhang/Disentangled-Multimodal-Learning.
Authors: Tianyi Zhang, Zheng-Peng Duan, Peng-Tao Jiang, Bo Li, Ming-Ming Cheng, Chun-Le Guo, Chongyi Li
Abstract: Diffusion-based real-world image super-resolution (Real-ISR) methods have demonstrated impressive performance.To achieve efficient Real-ISR, many works employ Variational Score Distillation (VSD) to distill pre-trained stable-diffusion (SD) model for one-step SR with a fixed timestep. However, since SD will perform different generative priors at different timesteps, a fixed timestep is difficult for these methods to fully leverage the generative priors in SD, leading to suboptimal performance.To address this, we propose a \textbf{T}ime-\textbf{A}ware one-step \textbf{D}iffusion Network for Real-ISR (\textbf{TADSR}). We first introduce a Time-Aware VAE Encoder, which projects the same image into different latent features based on timesteps.Through joint dynamic variation of timesteps and latent features, the student model can better align with the input pattern distribution of the pre-trained SD, thereby enabling more effective utilization of SD's generative capabilities.To better activate the generative prior of SD at different timesteps, we propose a Time-Aware VSD loss that bridges the timesteps of the student model and those of the teacher model, thereby producing more consistent generative prior guidance conditioned on timesteps. Additionally, though utilizing the generative prior in SD at different timesteps, our method can naturally achieve \textbf{controllable trade-offs between fidelity and realism} by changing the timestep.Experimental results demonstrate that our method achieves both state-of-the-art performance and controllable SR results with only a single step. The source codes are released at https://github.com/zty557/TADSR
Authors: Haruo Fujiwara, Yusuke Mukuta, Tatsuya Harada
Abstract: Recent advances in text-driven 3D scene editing and stylization, which leverage the powerful capabilities of 2D generative models, have demonstrated promising outcomes. However, challenges remain in ensuring high-quality stylization and view consistency simultaneously. Moreover, applying style consistently to different regions or objects in the scene with semantic correspondence is a challenging task. To address these limitations, we introduce techniques that enhance the quality of 3D stylization while maintaining view consistency and providing optional region-controlled style transfer. Our method achieves stylization by re-training an initial 3D representation using stylized multi-view 2D images of the source views. Therefore, ensuring both style consistency and view consistency of stylized multi-view images is crucial. We achieve this by extending the style-aligned depth-conditioned view generation framework, replacing the fully shared attention mechanism with a single reference-based attention-sharing mechanism, which effectively aligns style across different viewpoints. Additionally, inspired by recent 3D inpainting methods, we utilize a grid of multiple depth maps as a single-image reference to further strengthen view consistency among stylized images. Finally, we propose Multi-Region Importance-Weighted Sliced Wasserstein Distance Loss, allowing styles to be applied to distinct image regions using segmentation masks from off-the-shelf models. We demonstrate that this optional feature enhances the faithfulness of style transfer and enables the mixing of different styles across distinct regions of the scene. Experimental evaluations, both qualitative and quantitative, demonstrate that our pipeline effectively improves the results of text-driven 3D stylization. Project Page: https://haruolabs.github.io/improved-gs-style-page/
Authors: Weiqiao Han, Chenlin Meng, Christopher D. Manning, Stefano Ermon
Abstract: We present DistillKac, a fast image generator that uses the damped wave equation and its stochastic Kac representation to move probability mass at finite speed. In contrast to diffusion models whose reverse time velocities can become stiff and implicitly allow unbounded propagation speed, Kac dynamics enforce finite speed transport and yield globally bounded kinetic energy. Building on this structure, we introduce classifier-free guidance in velocity space that preserves square integrability under mild conditions. We then propose endpoint only distillation that trains a student to match a frozen teacher over long intervals. We prove a stability result that promotes supervision at the endpoints to closeness along the entire path. Experiments demonstrate DistillKac delivers high quality samples with very few function evaluations while retaining the numerical stability benefits of finite speed probability flows.
Authors: Matt Y. Cheung, Ashok Veeraraghavan, Guha Balakrishnan
Abstract: In clinical applications, the utility of segmentation models is often based on the accuracy of derived downstream metrics such as organ size, rather than by the pixel-level accuracy of the segmentation masks themselves. Thus, uncertainty quantification for such metrics is crucial for decision-making. Conformal prediction (CP) is a popular framework to derive such principled uncertainty guarantees, but applying CP naively to the final scalar metric is inefficient because it treats the complex, non-linear segmentation-to-metric pipeline as a black box. We introduce COMPASS, a practical framework that generates efficient, metric-based CP intervals for image segmentation models by leveraging the inductive biases of their underlying deep neural networks. COMPASS performs calibration directly in the model's representation space by perturbing intermediate features along low-dimensional subspaces maximally sensitive to the target metric. We prove that COMPASS achieves valid marginal coverage under the assumption of exchangeability. Empirically, we demonstrate that COMPASS produces significantly tighter intervals than traditional CP baselines on four medical image segmentation tasks for area estimation of skin lesions and anatomical structures. Furthermore, we show that leveraging learned internal features to estimate importance weights allows COMPASS to also recover target coverage under covariate shifts. COMPASS paves the way for practical, metric-based uncertainty quantification for medical image segmentation.
Authors: Jonas Ngnaw\'e, Maxime Heuillet, Sabyasachi Sahoo, Yann Pequignot, Ola Ahmad, Audrey Durand, Fr\'ed\'eric Precioso, Christian Gagn\'e
Abstract: Fine-tuning pretrained models is a standard and effective workflow in modern machine learning. However, robust fine-tuning (RFT), which aims to simultaneously achieve adaptation to a downstream task and robustness to adversarial examples, remains challenging. Despite the abundance of non-robust pretrained models in open-source repositories, their potential for RFT is less understood. We address this knowledge gap by systematically examining RFT from such non-robust models. Our experiments reveal that fine-tuning non-robust models with a robust objective, even under small perturbations, can lead to poor performance, a phenomenon that we dub suboptimal transfer. In challenging scenarios (eg, difficult tasks, high perturbation), the resulting performance can be so low that it may be considered a transfer failure. We find that fine-tuning using a robust objective impedes task adaptation at the beginning of training and eventually prevents optimal transfer. However, we propose a novel heuristic, Epsilon-Scheduling, a schedule over perturbation strength used during training that promotes optimal transfer. Additionally, we introduce expected robustness, a metric that captures performance across a range of perturbations, providing a more comprehensive evaluation of the accuracy-robustness trade-off for diverse models at test time. Extensive experiments on a wide range of configurations (six pretrained models and five datasets) show that Epsilon-Scheduling successfully prevents suboptimal transfer and consistently improves expected robustness.
Authors: Shu Liu, Wenlin Chen, Weihao Li, Zheng Wang, Lijin Yang, Jianing Huang, Yipin Zhang, Zhongzhan Huang, Ze Cheng, Hao Yang
Abstract: Diffusion-based planners have shown strong potential for autonomous driving by capturing multi-modal driving behaviors. A key challenge is how to effectively guide these models for safe and reactive planning in closed-loop settings, where the ego vehicle's actions influence future states. Recent work leverages typical expert driving behaviors (i.e., anchors) to guide diffusion planners but relies on a truncated diffusion schedule that introduces an asymmetry between the forward and denoising processes, diverging from the core principles of diffusion models. To address this, we introduce BridgeDrive, a novel anchor-guided diffusion bridge policy for closed-loop trajectory planning. Our approach formulates planning as a diffusion bridge that directly transforms coarse anchor trajectories into refined, context-aware plans, ensuring theoretical consistency between the forward and reverse processes. BridgeDrive is compatible with efficient ODE solvers, enabling real-time deployment. We achieve state-of-the-art performance on the Bench2Drive closed-loop evaluation benchmark, improving the success rate by 7.72% over prior arts. Project page: https://github.com/shuliu-ethz/BridgeDrive.
Authors: Xilong Zhou, Bao-Huy Nguyen, Lo\"ic Magne, Vladislav Golyanik, Thomas Leimk\"uhler, Christian Theobalt
Abstract: Radiance fields have emerged as a predominant representation for modeling 3D scene appearance. Neural formulations such as Neural Radiance Fields provide high expressivity but require costly ray marching for rendering, whereas primitive-based methods such as 3D Gaussian Splatting offer real-time efficiency through splatting, yet at the expense of representational power. Inspired by advances in both these directions, we introduce splattable neural primitives, a new volumetric representation that reconciles the expressivity of neural models with the efficiency of primitive-based splatting. Each primitive encodes a bounded neural density field parameterized by a shallow neural network. Our formulation admits an exact analytical solution for line integrals, enabling efficient computation of perspectively accurate splatting kernels. As a result, our representation supports integration along view rays without the need for costly ray marching. The primitives flexibly adapt to scene geometry and, being larger than prior analytic primitives, reduce the number required per scene. On novel-view synthesis benchmarks, our approach matches the quality and speed of 3D Gaussian Splatting while using $10\times$ fewer primitives and $6\times$ fewer parameters. These advantages arise directly from the representation itself, without reliance on complex control or adaptation frameworks. The project page is https://vcai.mpi-inf.mpg.de/projects/SplatNet/.
Authors: Heming Zou, Yunliang Zang, Wutong Xu, Xiangyang Ji
Abstract: Using a nearly-frozen pretrained model, the continual representation learning paradigm reframes parameter updates as a similarity-matching problem to mitigate catastrophic forgetting. However, directly leveraging pretrained features for downstream tasks often suffers from multicollinearity in the similarity-matching stage, and more advanced methods can be computationally prohibitive for real-time, low-latency applications. Inspired by the fly olfactory circuit, we propose Fly-CL, a bio-inspired framework compatible with a wide range of pretrained backbones. Fly-CL substantially reduces training time while achieving performance comparable to or exceeding that of current state-of-the-art methods. We theoretically show how Fly-CL progressively resolves multicollinearity, enabling more effective similarity matching with low time complexity. Extensive simulation experiments across diverse network architectures and data regimes validate Fly-CL's effectiveness in addressing this challenge through a biologically inspired design. Code is available at https://github.com/gfyddha/Fly-CL.
Authors: Jizhan Fang, Xinle Deng, Haoming Xu, Ziyan Jiang, Yuqi Tang, Ziwen Xu, Shumin Deng, Yunzhi Yao, Mengru Wang, Shuofei Qiao, Huajun Chen, Ningyu Zhang
Abstract: Despite their remarkable capabilities, Large Language Models (LLMs) struggle to effectively leverage historical interaction information in dynamic and complex environments. Memory systems enable LLMs to move beyond stateless interactions by introducing persistent information storage, retrieval, and utilization mechanisms. However, existing memory systems often introduce substantial time and computational overhead. To this end, we introduce a new memory system called LightMem, which strikes a balance between the performance and efficiency of memory systems. Inspired by the Atkinson-Shiffrin model of human memory, LightMem organizes memory into three complementary stages. First, cognition-inspired sensory memory rapidly filters irrelevant information through lightweight compression and groups information according to their topics. Next, topic-aware short-term memory consolidates these topic-based groups, organizing and summarizing content for more structured access. Finally, long-term memory with sleep-time update employs an offline procedure that decouples consolidation from online inference. On LongMemEval and LoCoMo, using GPT and Qwen backbones, LightMem consistently surpasses strong baselines, improving QA accuracy by up to 7.7% / 29.3%, reducing total token usage by up to 38x / 20.9x and API calls by up to 30x / 55.5x, while purely online test-time costs are even lower, achieving up to 106x / 117x token reduction and 159x / 310x fewer API calls. The code is available at https://github.com/zjunlp/LightMem.
Authors: Yupeng Xie, Zhiyang Zhang, Yifan Wu, Sirong Lu, Jiayi Zhang, Zhaoyang Yu, Jinlin Wang, Sirui Hong, Bang Liu, Chenglin Wu, Yuyu Luo
Abstract: Visualization, a domain-specific yet widely used form of imagery, is an effective way to turn complex datasets into intuitive insights, and its value depends on whether data are faithfully represented, clearly communicated, and aesthetically designed. However, evaluating visualization quality is challenging: unlike natural images, it requires simultaneous judgment across data encoding accuracy, information expressiveness, and visual aesthetics. Although multimodal large language models (MLLMs) have shown promising performance in aesthetic assessment of natural images, no systematic benchmark exists for measuring their capabilities in evaluating visualizations. To address this, we propose VisJudge-Bench, the first comprehensive benchmark for evaluating MLLMs' performance in assessing visualization aesthetics and quality. It contains 3,090 expert-annotated samples from real-world scenarios, covering single visualizations, multiple visualizations, and dashboards across 32 chart types. Systematic testing on this benchmark reveals that even the most advanced MLLMs (such as GPT-5) still exhibit significant gaps compared to human experts in judgment, with a Mean Absolute Error (MAE) of 0.553 and a correlation with human ratings of only 0.428. To address this issue, we propose VisJudge, a model specifically designed for visualization aesthetics and quality assessment. Experimental results demonstrate that VisJudge significantly narrows the gap with human judgment, reducing the MAE to 0.421 (a 23.9% reduction) and increasing the consistency with human experts to 0.687 (a 60.5% improvement) compared to GPT-5. The benchmark is available at https://github.com/HKUSTDial/VisJudgeBench.
Authors: Jintian Zhang, Kewei Xu, Jingsheng Zheng, Zhuoyun Yu, Yuqi Zhu, Yujie Luo, Lanning Wei, Shuofei Qiao, Lun Du, Da Zheng, Shumin Deng, Huajun Chen, Ningyu Zhang
Abstract: LLMs and Agents have achieved impressive progress in code generation, mathematical reasoning, and scientific discovery. However, existing benchmarks primarily measure correctness, overlooking the diversity of methods behind solutions. True innovation depends not only on producing correct answers but also on the originality of the approach. We present InnoGym, the first benchmark and framework designed to systematically evaluate the innovation potential of AI agents. InnoGym introduces two complementary metrics: performance gain, which measures improvement over the best-known solutions, and novelty, which captures methodological differences from prior approaches. The benchmark includes 18 carefully curated tasks from real-world engineering and scientific domains, each standardized through resource filtering, evaluator validation, and solution collection. In addition, we provide iGym, a unified execution environment for reproducible and long-horizon evaluations. Extensive experiments show that while some agents produce novel approaches, their lack of robustness limits performance gains. These results highlight a key gap between creativity and effectiveness, underscoring the need for benchmarks that evaluate both.
Authors: Sam Gijsen, Marc-Andre Schulz, Kerstin Ritter
Abstract: The development of foundation models for functional magnetic resonance imaging (fMRI) time series holds significant promise for predicting phenotypes related to disease and cognition. Current models, however, are often trained using a mask-and-reconstruct objective on small brain regions. This focus on low-level information leads to representations that are sensitive to noise and temporal fluctuations, necessitating extensive fine-tuning for downstream tasks. We introduce Brain-Semantoks, a self-supervised framework designed specifically to learn abstract representations of brain dynamics. Its architecture is built on two core innovations: a semantic tokenizer that aggregates noisy regional signals into robust tokens representing functional networks, and a self-distillation objective that enforces representational stability across time. We show that this objective is stabilized through a novel training curriculum, ensuring the model robustly learns meaningful features from low signal-to-noise time series. We demonstrate that learned representations enable strong performance on a variety of downstream tasks even when only using a linear probe. Furthermore, we provide comprehensive scaling analyses indicating more unlabeled data reliably results in out-of-distribution performance gains without domain adaptation.
Authors: Tianze Luo, Haotian Yuan, Zhuang Liu
Abstract: The multi-step denoising process in diffusion and Flow Matching models causes major efficiency issues, which motivates research on few-step generation. We present Solution Flow Models (SoFlow), a framework for one-step generation from scratch. By analyzing the relationship between the velocity function and the solution function of the velocity ordinary differential equation (ODE), we propose a Flow Matching loss and a solution consistency loss to train our models. The Flow Matching loss allows our models to provide estimated velocity fields for Classifier-Free Guidance (CFG) during training, which improves generation performance. Notably, our consistency loss does not require the calculation of the Jacobian-vector product (JVP), a common requirement in recent works that is not well-optimized in deep learning frameworks like PyTorch. Experimental results indicate that, when trained from scratch using the same Diffusion Transformer (DiT) architecture and an equal number of training epochs, our models achieve better FID-50K scores than MeanFlow models on the ImageNet 256x256 dataset.
Authors: Corentin Lobet, Francesca Chiaromonte
Abstract: Feature attribution is the dominant paradigm for explaining the predictions of complex machine learning models like neural networks. However, most existing methods offer little guarantee of reflecting the model's prediction-making process. We define the notion of explanatory alignment and argue that it is central to trustworthy predictive modeling: in short, it requires that explanations directly underlie predictions rather than serve as rationalizations. We present model readability as a design principle enabling alignment, and Pointwise-interpretable Networks (PiNets) as a modeling framework to pursue it in a deep learning context. PiNets combine statistical intelligence with a pseudo-linear structure that yields instance-wise linear predictions in an arbitrary feature space. We illustrate their use on image classification and segmentation tasks, demonstrating that PiNets produce explanations that are not only aligned by design but also faithful across other dimensions: meaningfulness, robustness, and sufficiency.
Authors: Rotem Gatenyo, Ohad Fried
Abstract: We study zero-shot 3D alignment of two given meshes, using a text prompt describing their spatial relation -- an essential capability for content creation and scene assembly. Earlier approaches primarily rely on geometric alignment procedures, while recent work leverages pretrained 2D diffusion models to model language-conditioned object-object spatial relationships. In contrast, we directly optimize the relative pose at test time, updating translation, rotation, and isotropic scale with CLIP-driven gradients via a differentiable renderer, without training a new model. Our framework augments language supervision with geometry-aware objectives: a variant of soft-Iterative Closest Point (ICP) term to encourage surface attachment and a penetration loss to discourage interpenetration. A phased schedule strengthens contact constraints over time, and camera control concentrates the optimization on the interaction region. To enable evaluation, we curate a benchmark containing diverse categories and relations, and compare against baselines. Our method outperforms all alternatives, yielding semantically faithful and physically plausible alignments.
Authors: Eslam Zaher, Maciej Trzaskowski, Quan Nguyen, Fred Roosta
Abstract: Latent-space optimization methods for counterfactual explanations - framed as minimal semantic perturbations that change model predictions - inherit the ambiguity of Wachter et al.'s objective: the choice of distance metric dictates whether perturbations are meaningful or adversarial. Existing approaches adopt flat or misaligned geometries, leading to off-manifold artifacts, semantic drift, or adversarial collapse. We introduce Perceptual Counterfactual Geodesics (PCG), a method that constructs counterfactuals by tracing geodesics under a perceptually Riemannian metric induced from robust vision features. This geometry aligns with human perception and penalizes brittle directions, enabling smooth, on-manifold, semantically valid transitions. Experiments on three vision datasets show that PCG outperforms baselines and reveals failure modes hidden under standard metrics.
Authors: Huanlin Gao, Ping Chen, Fuyuan Shi, Ruijia Wu, Li YanTao, Qiang Hui, Yuren You, Ting Lu, Chao Tan, Shaoan Zhao, Zhaoxiang Liu, Fang Zhao, Kai Wang, Shiguo Lian
Abstract: We present MeanCache, a training-free caching framework for efficient Flow Matching inference. Existing caching methods reduce redundant computation but typically rely on instantaneous velocity information (e.g., feature caching), which often leads to severe trajectory deviations and error accumulation under high acceleration ratios. MeanCache introduces an average-velocity perspective: by leveraging cached Jacobian--vector products (JVP) to construct interval average velocities from instantaneous velocities, it effectively mitigates local error accumulation. To further improve cache timing and JVP reuse stability, we develop a trajectory-stability scheduling strategy as a practical tool, employing a Peak-Suppressed Shortest Path under budget constraints to determine the schedule. Experiments on FLUX.1, Qwen-Image, and HunyuanVideo demonstrate that MeanCache achieves 4.12X and 4.56X and 3.59X acceleration, respectively, while consistently outperforming state-of-the-art caching baselines in generation quality. We believe this simple yet effective approach provides a new perspective for Flow Matching inference and will inspire further exploration of stability-driven acceleration in commercial-scale generative models.
Authors: Dung Anh Hoang, Cuong Pham anh Trung Le, Jianfei Cai, Thanh-Toan Do
Abstract: Diffusion models have shown remarkable performance in image synthesis by progressively estimating a smooth transition from a Gaussian distribution of noise to a real image. Unfortunately, their practical deployment is limited by slow inference speed, high memory usage, and the computational demands of the noise estimation process. Post-training quantization (PTQ) emerges as a promising solution to accelerate sampling and reduce memory overhead for diffusion models. Existing PTQ methods for diffusion models typically apply uniform weights to calibration samples across timesteps, which is sub-optimal since data at different timesteps may contribute differently to the diffusion process. Additionally, due to varying activation distributions and gradients across timesteps, a uniform quantization approach is sub-optimal. Each timestep requires a different gradient direction for optimal quantization, and treating them equally can lead to conflicting gradients that degrade performance. In this paper, we propose a novel PTQ method that addresses these challenges by assigning appropriate weights to calibration samples. Specifically, our approach learns to assign optimal weights to calibration samples to align the quantized model's gradients across timesteps, facilitating the quantization process. Extensive experiments on CIFAR-10, LSUN-Bedrooms, and ImageNet demonstrate the superiority of our method compared to other PTQ methods for diffusion models.
Authors: Yang Tian, Yu Fan, Liudmila Zavolokina, Sarah Ebling
Abstract: Text-to-image generative models have made remarkable progress in producing high-quality visual content from textual descriptions, yet concerns remain about how they represent social groups. While characteristics like gender and race have received increasing attention, disability representations remain underexplored. This study investigates how people with disabilities are represented in AI-generated images by analyzing outputs from Stable Diffusion XL and DALL-E 3 using a structured prompt design. We analyze disability representations by comparing image similarities between generic disability prompts and prompts referring to specific disability categories. Moreover, we evaluate how mitigation strategies influence disability portrayals, with a focus on assessing affective framing through sentiment polarity analysis, combining both automatic and human evaluation. Our findings reveal persistent representational imbalances and highlight the need for continuous evaluation and refinement of generative models to foster more diverse and inclusive portrayals of disability.
Authors: Bowen Ping, Chengyou Jia, Minnan Luo, Hangwei Qian, Ivor Tsang
Abstract: Reinforcement learning has emerged as a promising paradigm for aligning diffusion and flow-matching models with human preferences, yet practitioners face fragmented codebases, model-specific implementations, and engineering complexity. We introduce Flow-Factory, a unified framework that decouples algorithms, models, and rewards through through a modular, registry-based architecture. This design enables seamless integration of new algorithms and architectures, as demonstrated by our support for GRPO, DiffusionNFT, and AWM across Flux, Qwen-Image, and WAN video models. By minimizing implementation overhead, Flow-Factory empowers researchers to rapidly prototype and scale future innovations with ease. Flow-Factory provides production-ready memory optimization, flexible multi-reward training, and seamless distributed training support. The codebase is available at https://github.com/X-GenGroup/Flow-Factory.
Authors: Xiaoxi Li, Wenxiang Jiao, Jiarui Jin, Shijian Wang, Guanting Dong, Jiajie Jin, Hao Wang, Yinuo Wang, Ji-Rong Wen, Yuan Lu, Zhicheng Dou
Abstract: Human intelligence naturally intertwines omni-modal perception -- spanning vision, audio, and language -- with complex reasoning and tool usage to interact with the world. However, current multi-modal LLMs are primarily confined to bi-modal interactions (e.g., vision-language), lacking the unified cognitive capabilities required for general AI assistants. To bridge this gap, we introduce OmniGAIA, a comprehensive benchmark designed to evaluate omni-modal agents on tasks necessitating deep reasoning and multi-turn tool execution across video, audio, and image modalities. Constructed via a novel omni-modal event graph approach, OmniGAIA synthesizes complex, multi-hop queries derived from real-world data that require cross-modal reasoning and external tool integration. Furthermore, we propose OmniAtlas, a native omni-modal foundation agent under tool-integrated reasoning paradigm with active omni-modal perception. Trained on trajectories synthesized via a hindsight-guided tree exploration strategy and OmniDPO for fine-grained error correction, OmniAtlas effectively enhances the tool-use capabilities of existing open-source models. This work marks a step towards next-generation native omni-modal AI assistants for real-world scenarios.
Authors: Alaa Anani, Tobias Lorenz, Bernt Schiele, Mario Fritz, Jonas Fischer
Abstract: Understanding how neural networks arrive at their predictions is essential for debugging, auditing, and deployment. Mechanistic interpretability pursues this goal by identifying circuits - minimal subnetworks responsible for specific behaviors. However, existing circuit discovery methods are brittle: circuits depend strongly on the chosen concept dataset and often fail to transfer out-of-distribution, raising doubts whether they capture concept or dataset-specific artifacts. We introduce Certified Circuits, which provide provable stability guarantees for circuit discovery. Our framework wraps any black-box discovery algorithm with randomized data subsampling to certify that circuit component inclusion decisions are invariant to bounded edit-distance perturbations of the concept dataset. Unstable neurons are abstained from, yielding circuits that are more compact and more accurate. On ImageNet and OOD datasets, certified circuits achieve up to 91% higher accuracy while using 45% fewer neurons, and remain reliable where baselines degrade. Certified Circuits puts circuit discovery on formal ground by producing mechanistic explanations that are provably stable and better aligned with the target concept. Code will be released soon!
Authors: Gorkem Yildiz
Abstract: We present HELMLAB, a 72-parameter analytical color space for UI design systems. The forward transform maps CIE XYZ to a perceptually-organized Lab representation through learned matrices, per-channel power compression, Fourier hue correction, and embedded Helmholtz-Kohlrausch lightness adjustment. A post-pipeline neutral correction guarantees that achromatic colors map to a=b=0 (chroma < 10^-6), and a rigid rotation of the chromatic plane improves hue-angle alignment without affecting the distance metric, which is invariant under isometries. On the COMBVD dataset (3,813 color pairs), HELMLAB achieves a STRESS of 23.30, a 20.2% reduction from CIEDE2000 (29.18). A blue-band refit with sub-dataset penalties reduces gradient non-uniformity in the blue-cyan region by 8.9x at a cost of only +0.08 STRESS. Cross-validation on He et al. 2022 and MacAdam 1974 shows competitive cross-dataset performance. The transform is invertible with round-trip errors below 10^-14. Gamut mapping, design-token export, and dark/light mode adaptation utilities are included for use in web and mobile design systems.
Authors: Faiz Aladin, Ashwin Balasubramanian, Lars Lindemann, Daniel Seita
Abstract: Reachability analysis has become increasingly important in robotics to distinguish safe from unsafe states. Unfortunately, existing reachability and safety analysis methods often fall short, as they typically require known system dynamics or large datasets to estimate accurate system models, are computationally expensive, and assume full state information. A recent method, called MORALS, aims to address these shortcomings by using topological tools to estimate Regions of Attraction (ROA) in a low-dimensional latent space. However, MORALS still relies on full state knowledge and has not been studied when only sensor measurements are available. This paper presents Visual Morse Graph-Aided Estimation of Regions of Attraction in a Learned Latent Space (V-MORALS). V-MORALS takes in a dataset of image-based trajectories of a system under a given controller, and learns a latent space for reachability analysis. Using this learned latent space, our method is able to generate well-defined Morse Graphs, from which we can compute ROAs for various systems and controllers. V-MORALS provides capabilities similar to the original MORALS architecture without relying on state knowledge, and using only high-level sensor data. Our project website is at: https://v-morals.onrender.com.
Authors: Bin Xu, Yufei Zhou, Boling Song, Jingwen Sun, Yang Bian, Cheng Lu, Ye Wu, Jianfei Tu, Xiangxue Wang
Abstract: We propose a Hierarchical Multi-scale Knowledge-aware Graph Network (HMKGN) that models multi-scale interactions and spatially hierarchical relationships within whole-slide images (WSIs) for cancer prognostication. Unlike conventional attention-based MIL, which ignores spatial organization, or graph-based MIL, which relies on static handcrafted graphs, HMKGN enforces a hierarchical structure with spatial locality constraints, wherein local cellular-level dynamic graphs aggregate spatially proximate patches within each region of interest (ROI) and a global slide-level dynamic graph integrates ROI-level features into WSI-level representations. Moreover, multi-scale integration at the ROI level combines coarse contextual features from broader views with fine-grained structural representations from local patch-graph aggregation. We evaluate HMKGN on four TCGA cohorts (KIRC, LGG, PAAD, and STAD; N=513, 487, 138, and 370) for survival prediction. It consistently outperforms existing MIL-based models, yielding improved concordance indices (10.85% better) and statistically significant stratification of patient survival risk (log-rank p < 0.05).