Authors: Lv Tang, Tianyi Zheng, Yang Liu, Bo Li, Xingyu Li
Abstract: Visual text compression (VTC) promises efficient long-context processing by rendering text into an image and re-encoding it with a vision-language model, often producing $3$--$20\times$ fewer decoder tokens than subword tokenization. Yet token savings do not translate predictably into downstream utility: on some tasks the visual path matches or exceeds the text path, on others it collapses, and the compression ratio itself does not predict which regime will occur. The missing quantity is therefore not another summary of efficiency, but a principled measure of task-relevant information loss induced by visual encoding. We address this problem by formulating VTC in the language of measure transport. Treating text and visual tokens as empirical probability measures, we show that the ViT patch encoder induces a push-forward map whose transport cost decomposes into a precision cost from within-patch aggregation and a coverage cost from cross-patch fragmentation. Both terms are estimable from downstream-label-free probes. This formulation yields two operational consequences: a downstream-label-free routing criterion that selects whether to use the visual path for a given input or benchmark instance, and a transport-informed foveation mechanism that re-encodes high-cost regions at higher resolution. Across $24$ NLP datasets at Qwen3-4B, our label-free rule matches the per-dataset oracle on $17/24$ datasets ($70.8\%$), and improves the average task score by $+3.3\%$ with $-10.3\%$ average tokens relative to a pure-LLM.
Authors: Yiwen Xu, Tariq M. Khan, Yang Song, Erik Meijering
Abstract: Edge deep learning, a paradigm change reconciling edge computing and deep learning, facilitates real-time decision making attuned to environmental factors through the close integration of computational resources and data sources. Here we provide a comprehensive review of the current state of the art in edge deep learning, focusing on computer vision applications, in particular medical diagnostics. An overview of the foundational principles and technical advantages of edge deep learning is presented, emphasising the capacity of this technology to revolutionise a wide range of domains. Furthermore, we present a novel categorisation of edge hardware platforms based on performance and usage scenarios, facilitating platform selection and operational effectiveness. Following this, we dive into approaches to effectively implement deep neural networks on edge devices, encompassing methods such as lightweight design and model compression. Reviewing practical applications in the fields of computer vision in general and medical diagnostics in particular, we demonstrate the profound impact edge-deployed deep learning models can have in real-life situations. Finally, we provide an analysis of potential future directions and obstacles to the adoption of edge deep learning, with the intention to stimulate further investigations and advancements of intelligent edge deep learning solutions. This survey provides researchers and practitioners with a comprehensive reference shedding light on the critical role deep learning plays in the advancement of edge computing applications.
Authors: Yufan Deng, Daquan Zhou
Abstract: Progress in embodied intelligence increasingly depends on scalable data infrastructure. While vision and language have scaled with internet corpora, learning physical interaction remains constrained by the lack of large, diverse, and richly annotated human activity data. We present HumanNet, a one-million-hour human-centric video corpus that captures how humans interact with the physical world at scale. HumanNet spans both first-person and third-person perspectives and covers fine-grained activities, human-object interactions, tool use, and long-horizon behaviors across diverse real-world environments. Beyond raw video, the dataset provides interaction-centric annotations, including captions, motion descriptions, and hand and body-related signals, enabling motion-aware and interaction-aware learning. Beyond scale, HumanNet introduces a systematic data curation paradigm for embodied learning, where human-centric filtering, temporal structuring, viewpoint diversity, and annotation enrichment are treated as first-class design principles. This design transforms unstructured internet video into a scalable substrate for representation learning, activity understanding, motion generation, and human-to-robot transfer. We conduct a first-step validation on the value of this design through controlled vision-language-action ablation: under a fixed set of validation data, continued training from the Qwen VLM model with 1000 hours of egocentric video drawn from HumanNet surpasses the continued training with 100 hours of real-robot data from Magic Cobot, indicating that egocentric human video could be a scalable and cost-effective substitute for robot data. By building this project, we aim to explore the opportunity to scale embodied foundation models using human-centric videos, rather than relying solely on robot-specific data.
Authors: Zhifeng Gu, Yuqi Wang, Bing Wang
Abstract: Relative spatial relations provide a compact representation of spatial structure and are fundamental to relative spatial reasoning in 3D layout generation. Recent works leverage Multimodal Large Language Models (MLLMs) to infer such relations, but the inferred relations are often unreliable and are typically handled with post-hoc heuristics. In this paper, we propose R$^3$L, a general framework that improves the reliability and consistency of relative spatial reasoning for 3D layout generation. Our key motivation is that multi-hop reasoning requires repeated reference-frame transformations, which accumulate errors in inferred relations and lead to semantic and metric drift. To mitigate this, we propose invariant spatial decomposition to break coupled relation chains, and consistent spatial imagination to promote self-consistency through an imagine-and-revise loop. We further introduce supportive spatial optimization to ease pose optimization via global-to-local coordinate re-parameterization. Extensive experiments across diverse scene types and instructions demonstrate that R$^3$L produces more physically feasible and semantically consistent layouts. Notably, our analysis shows that resolving frame-induced inconsistencies is crucial for reliable multi-hop relative spatial reasoning. The code is available at https://github.com/Neal2020GitHub/R3L.
Authors: Ali Salamatian, Anthony Fuller, Pritam Sarkar, James R. Green, Leonid Sigal, Evan Shelhamer
Abstract: Transformers dominate video recognition. They split videos into tokens, and processing them has expensive superlinear computational cost. Yet videos are filled with redundancy, so we can question the need for this expense. We introduce LookWhen, a selector-extractor framework that factorizes video recognition into learning when, where, and what to compute. Our shallow selector gets a scaled-down video and quickly scores all tokens across space-time, while our deep extractor gets the top-K selected tokens to approximate full-video representations without actually processing all the tokens. A key challenge is defining effective supervision for selection and extraction. For selection pre-training, we introduce a score on representations that ranks tokens by uniqueness using a simple nearest-neighbor distance. For extraction pre-training, we distill both a video teacher and an image teacher, for which we normalize its frame-wise representations to learn what changes within videos. Through these strategies, our selector-extractor learns general and efficient representations for feature extraction or fine-tuning to a task. Through experiments on Kinetics-400, SSv2, Epic-Kitchens, Diving48, Jester, and Charades, we show that LookWhen achieves a better accuracy-computation trade-off than efficient models and upgraded baselines of similar size. LookWhen Pareto-dominates in accuracy-FLOPs on 9 of 12 cases (6 tasks x 2 settings) and roughly matches on 3. In accuracy-throughput, measuring time in practice, LookWhen is more efficient still at 6.7x faster than InternVideo2-B at equal accuracy.
Authors: Ho Hin Lee, Dongna Du, Chu Wang, Yuankai Huo, Shi Gu, James C. Gee, Yifan Wu
Abstract: Vision foundation models are increasingly moving beyond 2D to volumetric domains such as 3D medical imaging, where unified pretraining across different imaging modalities (i.e. CT, MRI, and PET) could provide foundational models for diverse clinical tasks. However, training such models requires mixing heterogeneous imaging domains, and current mixture strategies remain largely heuristic. In this work, we observe that different medical imaging domains scale at variable rates during pretraining, and knowledge transfer between domains is strongly asymmetric: training on one domain can substantially improve another, but the reverse may be much weaker. Interestingly, both MAE reconstruction loss and cross-domain transfer follow predictable power-law trends with domain-specific behaviors. Motivated by these findings, we formulate data allocation as a scaling-law optimization problem. The derived allocations reveal an interpretable hub-and-island structure: highly transferable domains emerge as hubs that benefit many others and deserve strategic allocation, while isolated domains act as islands requiring direct investment. Empirically, transfer-aware allocation outperforms data-proportional sampling by up to 58% and generalizes well to unseen budgets with r=0.989. Downstream validation on disease classification and organ/lesion segmentation further confirms that the derived transfer-aware mixtures provide stronger pretrained representations for clinical 3D medical imaging tasks.
Authors: Yongjae Lee, Jingxing Li, Abhay Kumar Yadav, Rama Chellappa, Deliang Fan
Abstract: Adaptive density control in 3D Gaussian Splatting (3DGS) repeatedly grows the Gaussian population through fixed-cardinality random splitting to discover useful scene structure. However, in vanilla 3DGS, its binary split operator requires many densification rounds to expose fine details, making it a bottleneck for efficient training schedules with fewer iterations. We introduce AdpSplit, an error-driven adaptive split operator that determines the number of split children and initializes the child parameters from L1-pixel-error region statistics, enabling fewer densification iterations, thus reduced training time, while preserving the rendering quality of full-schedule training. Across the MipNeRF360, Deep-Blending, and Tanks&Temples datasets, AdpSplit reduces the training time of multiple accelerated 3DGS pipelines by 9.2%-22.3% as a simple drop-in replacement for the standard split operator. With FastGS, AdpSplit matches the full-schedule PSNR on MipNeRF360 while reducing training time by 16.4%, corresponding to a 12.6x acceleration over vanilla 3DGS.
Authors: Francisco Chen, Yiran Wang, Yunpeng Shi
Abstract: Pairwise translation directions are a key input to camera location estimation in global structure-from-motion. Existing estimators usually process each image pair independently, producing directions that may be locally plausible but inconsistent with the other relative directions in the viewing graph. To jointly estimate the direction, we propose TriDE, which exploits camera-triangle consistency as an efficient higher-order verification signal. Instead of solving a costly global nonlinear optimization problem that is sensitive to initialization, TriDE refines unreliable pairwise directions through message passing between directions and their incident weighted triangles. This information propagation strategy enables us to establish a strong phase-transition bound for exact recovery under a realistic random corruption model. Experiments on real image graphs show that TriDE improves direction accuracy by a large margin and yields better downstream camera locations, providing a practical link between local pairwise estimation and global camera pose geometry.
Authors: Aditya Parikh, Stella Frank, Sneha Das, Aasa Feragen
Abstract: Labeled datasets reflect the biases of their annotation pipelines, which sometimes introduce label bias: group-conditional label errors that cause systematic performance disparities across demographic subgroups. Label bias in image segmentation remains underexplored, as even detecting it typically requires clean, unbiased annotations, which are not readily available. We present a data-centric adaptation of Confident Learning to segmentation, allowing detection of label bias directly in the training data without a clean, unbiased ground truth. By comparing the provided training labels to the model's confident predictions, we isolate directional errors that quantify the presence and nature of bias, where standard overlap metrics like Dice fail. We further show that label bias influences subgroup separability in the encoder's feature space, an artifact we leverage for bias mitigation rather than suppressing it. We evaluate three datasets, spanning from synthetic to real-life bias, showing how our framework reliably detects and mitigates bias without access to clean labels, achieving equitable performance across experimental conditions.
Authors: Ernie Chu, Vishal M. Patel
Abstract: Diffusion Transformers (DiTs) have achieved state-of-the-art video generation quality, but they incur immense computational cost because standard inference applies the same number of denoising steps uniformly to every token in the sequence. It is well known that human vision ignores vast amounts of redundant motion. Why, then, do our densest models treat every spatiotemporal token with equal priority? In this paper, we introduce Heterogeneous Step Allocation (HSA), a training-free inference algorithm that assigns varying step budgets to different spatiotemporal tokens based on their velocity dynamics. To resolve the resulting sequence-length mismatch without sacrificing global context, HSA introduces a KV-cache synchronization mechanism that allows active tokens to attend to the full sequence while entirely bypassing inactive tokens. Furthermore, we derive a cached Euler update that advances the latent states of skipped tokens in a single operation without additional model evaluations. We evaluate HSA on the Wan-2 and LTX-2 models for both text-to-video (T2V) and image-to-video (I2V) generation. Our results demonstrate that HSA significantly outperforms previous state-of-the-art caching methods and the vanilla Flow Matching baseline, especially at aggressive acceleration regimes (e.g., 50% and 25% runtimes). Crucially, HSA achieves a superior quality-runtime Pareto frontier without the need for expensive offline profiling, robustly preserving structural integrity and generation quality even under tight computational budgets. Project page: https://ernestchu.github.io/hsa
Authors: Kirill Trapeznikov, Gabriel Mancino-Ball, Jonathan Li, Paul Cummer, Jai Aslam, Danial Samadi Vahdati, Tai Nguyen, Matthew C. Stamm, Peter Bautista, Michael Davinroy, Laura Cassani, Jill Crisman
Abstract: The proliferation of generative video technologies has intensified the need for reliable methods to detect and characterize synthetic media. To address this challenge, we organized the \href{https://safe-video-2025.dsri.org}{SAFE: Synthetic Video Detection Challenge}, co-located with the \textit{Authenticity and Provenance in the Age of Generative AI (APAI) Workshop }at ICCV 2025. The competition invited participants to develop and evaluate algorithms capable of distinguishing real from synthetic videos under fully blind evaluation conditions with over 600 submissions from 12 teams over a 90 day span. Hosted on the Hugging Face platform, the challenge comprised two primary tasks: (1) detection of synthetic video content generated by diverse state-of-the-art models, and (2) detection of synthetic content following common post-processing operations such as resizing, re-compression, motion blur and others. The challenge data consisted of 13 modern high quality synthetic video models with generated content matched to real videos from 21 diverse and challenge sources, all adding up to 20 hours of 6,000 video samples. This paper describes the challenge design, dataset construction, evaluation methodology, and outcomes, offering insights into the generalization and robustness of contemporary synthetic video detection methods. Our findings highlight measurable progress in cross-generator generalization but also persistent vulnerabilities to post-processing artifacts. https://safe-video-2025.dsri.org
URLs: https://safe-video-2025.dsri.org, https://safe-video-2025.dsri.org
Authors: Do Xuan Long, Yale Song, Min-Yen Kan, Tomas Pfister, Long T. Le
Abstract: Synthesizing consistent and coherent long video remains a fundamental challenge. Existing methods suffer from semantic drift and narrative collapse over long horizons. We present A$^2$RD, an Agentic Auto-Regressive Diffusion architecture that decouples creative synthesis from consistency enforcement. A$^2$RD formulates long video synthesis as a closed-loop process that synthesizes and self-improves video segment-by-segment through a Retrieve--Synthesize--Refine--Update cycle. It comprises three core components: (i) Multimodal Video Memory that tracks video progression across modalities; (ii) Adaptive Segment Generation that switches among generation modes for natural progression and visual consistency; and (iii) Hierarchical Test-Time Self-Improvement that self-improves each segment at frame and video levels to prevent error propagation. We further introduce LVBench-C, a challenging benchmark with non-linear entity and environment transitions to stress-test long-horizon consistency. Across public and LVBench-C benchmarks spanning one- to ten-minute videos, A$^2$RD outperforms state-of-the-art baselines by up to 30% in consistency and 20% in narrative coherence. Human evaluations corroborate these gains while also highlighting notable improvements in motion and transition smoothness.
Authors: Tony Tran, Richie R. Suganda, Bin Hu
Abstract: Object detection on heterogeneous edge devices must satisfy strict energy, latency, and memory constraints while still providing reliable perception for downstream autonomy. Existing energy-aware NAS methods often target limited deployment settings, while real energy remains difficult to optimize because it is highly device-dependent and costly to measure. We address these challenges with an energy-adaptive framework that combines an energy-aware XiResOFA search space, a two-stage energy estimator, and iterative search to identify a single energy-efficient base architecture. We then apply compound scaling to transform this base design into the XiYOLO family across deployment budgets, enabling interpretable accuracy-energy tradeoffs under sparse hardware measurements. Experiments on PascalVOC, COCO, and real-device deployment show that XiYOLO achieves a stronger energy-accuracy tradeoff than YOLO baselines. On PascalVOC, the medium XiYOLO model reaches 86.15 mAP50 while reducing energy relative to YOLOv12m by 20.6% on GPU and 35.9% on NPU. On COCO, XiYOLO reduces energy relative to YOLOv12 by up to 53.7% on GPU and 51.6% on NPU at the small scale. The proposed two-stage estimator also improves sample efficiency over a joint predictor under few-shot adaptation with only 2-20 target-device samples.
Authors: Yuchen Guo, Junli Gong, Yao Lu, Xintong Xu, Yiuming Cheung, Weifeng Su
Abstract: Infrared-Visible image fusion (IVIF) aims to integrate thermal information and detailed spatial structures into a single fused image to enhance perception. However, existing evaluation approaches tend to over-optimize both hand-crafted no-reference statistics and full-reference metrics that treat the source images as pseudo ground truths. Recent IVIF reward-modelling efforts learn from human ratings but use scalar regression on aggregated scores, neither leveraging the reasoning of Multimodal Large Language Models (MLLMs) nor encoding per-image perceptual ambiguity in their supervision, but naively introducing MLLMs with discrete one-hot supervision likewise collapses fused images of similar quality into different rating levels. To address this, we introduce FuScore, which utilizes an MLLM to mimic human visual perception by producing continuous quality score, rather than discrete level predictions, enabling fine-grained discrimination among fused images of similar quality. We exploit the agreement among four IVIF-specific sub-dimensions to construct a per-image soft label whose sharpness reflects how consensual the overall judgment is. We further introduce a tripartite objective combining per-image distributional supervision, within-source-pair Thurstone fidelity for method-level ordering, and cross-source-pair Thurstone fidelity for scene-level ordering across scenes. Extensive experiments demonstrate that FuScore achieves state-of-the-art correlation with human visual preferences.
Authors: Maria Despoina Siampou, Gengchen Mai, Ni Lao, Jinmeng Rao, Neha Arora, Cyrus Shahabi, Shushman Choudhury
Abstract: Multimodal self-supervised learning (MSSL) has emerged as a key paradigm for pretraining geospatial foundation models. However, existing geospatial MSSL methods are mainly designed for static pairs of modalities, such as satellite imagery, street-view imagery, and text, where learning is driven by aligning observations from the same or nearby locations. This assumption breaks down for human mobility trajectories, which represent continuous movement along paths rather than discrete observations at individual locations. Although trajectories are important for urban understanding through their ability to capture human activity across roads, neighborhoods, and places over time, they remain largely underexplored in current geospatial MSSL frameworks. We present TrajGANR, a novel trajectory-centric geospatial MSSL framework that aligns continuous movement patterns with static, location-based observations. TrajGANR learns a continuous neural representation of trajectories at arbitrary points along each path, which enables fine-grained alignment with nearby street-view images, even when they are not co-located with any trajectory waypoints. We leverage this capability to introduce an MSSL objective that jointly aligns three modalities: trajectories, street-view images, and their geographic locations. We evaluate TrajGANR on four urban mobility and road understanding tasks. Across these tasks, TrajGANR consistently outperforms existing geospatial MSSL frameworks and a trajectory-specific foundation model. Ablation studies further demonstrate that our proposed MSSL objective and the multimodal learning framework are the primary drivers of these improvements, highlighting the importance of fine-grained geospatial alignment over coarser aggregation, as well as geospatial multimodal learning.
Authors: Roy Xie, Dan Friedman, Donghan Yu, Bowen Pan, Christopher Fifty, Jang-Hyun Kim, Xianzhi Du, Zhe Gan, Vivek Rathod, Bhuwan Dhingra
Abstract: Vision Language Models (VLMs) offer the exciting possibility of processing text as rendered images, bypassing the need for tokenizing the text into long token sequences. Since VLM image encoders map fixed-size images to a fixed number of visual tokens, varying rendering resolution provides a fine-grained compression knob. However, accuracy deteriorates quickly as compression increases: characters shrink below the vision encoder's effective resolution, making them indistinguishable. To address this, we propose LensVLM, an inference framework and post-training recipe that enables VLMs to scan compressed images, then selectively expand only the relevant images to their uncompressed form via learned tools. Building on Qwen3.5-9B-Base, LensVLM maintains accuracy comparable to the full-text upper bound at 4.3x effective compression and outperforms retrieval-based, text- and visual-compression baselines up to 10.1x effective compression across seven text QA benchmarks. LensVLM also generalizes to multimodal document and code understanding tasks, with the accuracy gain over baselines growing as compression increases. Our analysis validates this approach: training makes visual compression robust to rendering choices, and as compression grows the model increasingly relies on expanded content rather than unreliable visual reading. The analysis also yields practical tool-choice guidance: text expansion is preferable for rendered text, while high-resolution image expansion suits native documents whose layout cues carry task-relevant information.
Authors: Yang Luo, Yan Gong, Yongsheng Gao, Jie Zhao, Xinyu Zhang, Huaping Liu
Abstract: In many practical 6D object pose estimation scenarios, we often have access to only a single real-world RGB-D reference view per object, typically without CAD models. Existing methods largely rely on explicit 3D models or multi-view data, which limits their scalability. To address this challenging single-reference model-free setting, we propose \textbf{OneViewAll}, a semantic-prior-guided framework that performs pose estimation via a novel Project-and-Compare paradigm. Instead of relying on computationally expensive CAD-based rendering, our method directly aligns reference and query observations within a projection-equivariant space. OneViewAll progressively integrates hierarchical semantic priors across three levels: (1) \textit{category- and scene-level} priors for efficient hypothesis initialization; (2) \textit{object-level symmetry} priors for geometry completion via mirror fusion; and (3) \textit{patch-level} priors for discriminative refinement. Extensive experiments demonstrate that OneViewAll achieves \textbf{92.5\%} ADD-0.1 accuracy on the LINEMOD dataset using only one real reference view -- significantly outperforming the CVPR 2025 baseline One2Any (52.6\%). It also yields consistent improvements on YCB-V, Real275, and Toyota-Light while maintaining low inference latency. Our results underscore the efficacy of symmetry-aware projection in handling symmetric, texture-less, and occluded objects.
Authors: Qiangqiang Wu, Grace McIlvain, Zhou Yu, Junhao Wen
Abstract: Foundation models (FMs) have shown great promise in medical imaging, but most FMs are trained on unimodal data within isolated domains, such as brain MRI alone. Human aging and disease arise through coordinated biological processes across organs, therefore motivating multimodal FMs that learn whole-body representations. A key challenge, however, is that real-world multimodal biomedical data are often missing not at random, which can reduce power, limit generalizability, and introduce bias. We propose Pan-FM, a pan-organ foundation model pre-trained on imaging from seven organs (Brain, Heart, Adipose, Liver, Kidney, Spleen, and Pancreas) under realistic missing-organ scenarios. Pan-FM uses a unified backbone that handles organ missingness during both training and inference, and is pre-trained with masking-based self-distillation. We find that naive multimodal pre-training leads to dominant-organ shortcut learning bias, with the model over-relying on dominant organs such as adipose and heart. To address this, we introduce Saliency-Guided Masking (SGM), which uses the model attention distribution to adaptively mask dominant organs during pre-training, thus encouraging more balanced cross-organ, whole-body learning. Notably, SGM introduces negligible computational overhead and can be seamlessly integrated into existing self-supervised learning frameworks to improve multi-organ representation learning. On the UK Biobank, Pan-FM achieves stronger prediction across 13 disease categories and 14 single disease entities than single-organ and multi-organ baselines, with improved robustness under missing-organ settings. Pan-FM serves as a scalable solution to realistic modality-missingness in multimodal learning in system neuroscience and as a step toward more generalizable whole-body FMs.
Authors: Yaozong Zheng, Bineng Zhong, Qihua Liang, Shuimu Zeng, Haiying Xia, Shuxiang Song
Abstract: How to achieve vision-language (VL) tracking using natural language descriptions from a video sequence \textbf{without relying on any bounding-box ground truth}? In this work, we achieve this goal by tackling \textit{self-supervised VL tracking}, which aims to evaluate tracking capabilities guided by natural language descriptions. We introduce \textbf{\tracker}, a novel self-supervised VL tracker that is capable of tracking any referred object by a language description. Unlike traditional methods that equally fuse all language and visual tokens, we propose an efficient Dynamic Token Aggregation Module, which treats each visual token \textbf{unequally}. The module consists of three main steps: i) Based on an anchor token, it selects multiple important target tokens from the template frame. ii) The selected target tokens are merged according to their attention scores and aggregated into the language tokens, thereby eliminating redundant visual token noise and enhancing semantic alignment. iii) Finally, the fused language tokens serve as guiding signals to extract potential target tokens from the search frame and propagate them to subsequent frames, enhancing temporal prompts and encouraging the tracker to autonomously learn instance tracking from unlabeled videos. This new modeling approach enables the effective self-supervised learning of language-guided tracking representations without the need for large-scale bounding box annotations. Extensive experiments on VL tracking benchmarks show that {\tracker} surpasses SOTA self-supervised methods.
Authors: Zhiyuan Wang (Hefei University of Technology), Yanxiang Chen (Key Laboratory of Knowledge Engineering with Big Data), Yuanzhi Yao (Key Laboratory of Knowledge Engineering with Big Data), Yunfeng Diao (Key Laboratory of Knowledge Engineering with Big Data)
Abstract: Detecting AI-generated images across unseen architectures remains challenging, as existing models often overfit to generator-specific fingerprints and semantic content rather than learning universal forgery traces. We attribute this failure to feature entanglement: detectors learn these factors as a single entangled representation, where universal forgery traces are inextricably confounded with both generator-specific fingerprints and semantic content. Crucially, our spectral analysis reveals that this entanglement is avoidable: distinct generator-specific fingerprints (e.g., GAN stripes vs. Diffusion Model spots) occupy disjoint frequency subspaces and coexist as independent superpositions. Leveraging this physical orthogonality, we propose the Orthogonal Decomposition and Purification Network (ODP-Net) to structurally disentangle these factors. Specifically, ODP-Net employs (1) Instance-aware Orthogonal Decomposition to project features into mutually exclusive subspaces: universal forgery traces, generator-specific fingerprints, and semantic content; (2) Perturbation-based Purification to enforce semantic invariance via cross-sample feature injection; and (3) Manifold Alignment to bridge domain gaps. By explicitly decoupling universal forgery traces from generator-specific fingerprints and semantic content, ODP-Net achieves state-of-the-art performance on unseen architectures (e.g., Stable Diffusion 3), validating that structural disentanglement is key to generalization.
Authors: Xinyu Zhang, Zhengtong Xu, Yutian Tao, Yeping Wang, Yu She, Abdeslam Boularias
Abstract: World models predict future transitions from observations and actions. Existing works predominantly focus on image generation only. Visual feature-based world models, on the other hand, predict future visual features instead of raw video pixels, offering a promising alternative that is more efficient and less prone to hallucination. However, current feature-based approaches rely on direct regression, which leads to blurry or collapsed predictions in complex interactions, while generative modeling in high-dimensional feature spaces still remains challenging. In this work, we discover that a new type of latent action representation, which we refer to as *Residual Latent Action* (RLA), can be easily learned from DINO residuals. We also show that RLA is predictive, generalizable, and encodes temporal progression. Building on RLA, we propose *RLA World Model* (RLA-WM), which predicts RLA values via flow matching. RLA-WM outperforms both state-of-the-art feature-based and video-diffusion world models on simulation and real-world datasets, while being orders of magnitude faster than video diffusion. Furthermore, we develop two robot learning techniques that use RLA-WM to improve policy learning. The first one is a minimalist world action model with RLA that learns from actionless demonstration videos. The second one is the first visual RL framework trained entirely inside a world model learned from offline videos only, using a video-aligned reward and no online interactions or handcrafted rewards. Project page: https://mlzxy.github.io/rla-wm
Authors: Xinquan Yang, Congmin Wang, Xuguang Li, Yulei Li, Linlin Shen, Yongqiang Deng He Meng
Abstract: In the design of surgical guides for implant placement, determining the precise implant position is a critical step. However, the implant region itself is often characterized by a lack of distinctive texture in medical images. Consequently, artificial intelligence (AI) models must infer the correct implant position and angulation (slope) primarily by analyzing the texture of the surrounding teeth, which poses a significant challenge. To address this, we propose ImplantMamba, a network architecture designed for long-range sequential modeling to integrate texture information from adjacent teeth. Our approach explicitly couples the regression of the implant position with its slope. The core of ImplantMamba is a hybrid encoder that combines Convolutional Neural Networks (CNNs) with Mamba layers. This design enables the network to hierarchically extract local anatomical features through CNNs while simultaneously modeling global contextual dependencies across the entire scan volume via Mamba's selective scan operations, leading to a more comprehensive understanding of the implant site. Furthermore, we introduce a Slope-Coupled Prediction Branch (SCP). This branch is designed to connect the prediction of implant position with the slope, ensuring internal consistency and anatomical plausibility by thereby enforcing a coherent relationship between the predicted implant location and its angulation. Extensive experiments on a large-scale dental implant dataset demonstrate that the proposed ImplantMamba achieves superior performance compared to existing methods.
Authors: Houman Safaai, Andrew T. Landau, Celia C. Beron, Yasin Mazloumi, Bernardo L. Sabatini
Abstract: Channel importance in vision networks is usually summarized by a single score. That summary hides two different questions: how much a channel is related to the task, and whether its function can be supplied by same-layer peers when the channel is removed. We call the second property local replaceability. We introduce a two-axis view that separates these questions. The local axis measures input capture and peer overlap, while the target axis measures task information and target-excess information. Across ResNet-18, VGG-16, and MobileNetV2 trained on CIFAR-100, the two axes are weakly aligned, induce different channel groupings, and separate rapidly during training despite being strongly coupled at random initialization. A Gaussian linear analysis accounts for how this separation can arise through residualized gradient directions, and lesion plus peer-replacement experiments show that peer support refines removability beyond input capture and task relevance alone. Under the fixed FLOPs-matched pruning protocol, local-axis metrics are more reliable predictors of removability than target-axis metrics across the three CIFAR-100 backbones, with the same direction preserved in stress tests on CIFAR-10, Tiny-ImageNet, ImageNet-100, and a ConvNeXt-T/ImageNet-100 pilot. These findings identify an axis-level distinction rather than a universal ranking of pruning scores: local replaceability is a more reliable guide to removability than target relevance, while norm-based baselines remain competitive in architectures such as VGG-16. Relevance-based scores ask what a channel says about the task; pruning asks whether the network still needs that channel when its peers remain available.
Authors: Hongyang Zhang, Maonnan Wang, Ziyao Wang, Hongrui Yin, Man OnPun
Abstract: Cross-view geo-localization (CVGL) is fundamental for precise localization and navigation in GPS-denied environments, aiming to match ground or UAV imagery with satellite views. While existing approaches rely on global feature alignment, they often suffer from substantial domain shifts induced by varying regional textures and weather conditions. This issue becomes even more pronounced in UAV-based scenarios, where the broader perspective inevitably introduces dense, fine-grained objects, creating significant visual clutter. To address this, we draw inspiration from Object-Centric Learning (OCL) and propose InfoGeo, an information-theoretic framework designed to enhance robustness and generalization. InfoGeo reformulates the optimization as an information bottleneck process with two core objectives: (i) maximizing view-invariant information by aligning the object-centric structural relations across views, and (ii) minimizing view-specific noisy signals through cross-view knowledge constraints. Extensive evaluations across diverse benchmarks and challenging scenarios demonstrate that InfoGeo significantly outperforms state-of-the-art methods.
Authors: Talha Ilyas, Deval Mehta, Zongyuan Ge
Abstract: Skeleton-based human activity recognition has achieved strong empirical performance, yet most existing models remain black boxes and difficult to interpret. In this work, we introduce a neurosymbolic formulation of skeleton-based HAR that reframes action recognition as concept-driven first-order logical reasoning over motion primitives. Our framework bridges representation learning and symbolic inference by grounding first-order logic predicates in learnable spatial and temporal motion concepts. Specifically, we employ a standard spatio-temporal skeleton encoder to extract latent motion representations, which are then mapped to interpretable concept predicates via a spatio-temporal concept decoder that explicitly separates pose-centric and dynamics-centric abstractions. These concept predicates are composed through differentiable first-order logic layers, enabling the model to learn human-readable logical rules that govern action semantics. To impose semantic structure on the learned concepts, we align skeleton representations with LLM-derived descriptions of atomic motion primitives, establishing a shared conceptual space for perception and reasoning. Extensive experiments on NTU RGB+D 60/120 and NW-UCLA demonstrate that our approach achieves competitive recognition performance while providing explicit, interpretable explanations grounded in logical structure. Our results highlight neurosymbolic reasoning as an effective paradigm for interpretable spatio-temporal action understanding. Code: https://github.com/Mr-TalhaIlyas/REASON
Authors: Yuan Yao, Qiushi Yang, Humen Zhong, Jiangning Wei, Yifang Men, Shuai Bai, Miaomiao Cui, Zhibo Yang
Abstract: Open-world referring segmentation requires grounding unconstrained language expressions to precise pixel-level regions. Existing multimodal large language models (MLLMs) exhibit strong open-world visual grounding, but their outputs remain limited to sparse bounding-box coordinates and are insufficient for dense visual prediction. Recent MLLM-based segmentation methods either directly predict sparse contour coordinates, struggling to reconstruct continuous object boundaries, or rely on external segmentation foundation models such as the Segment Anything Model (SAM), introducing substantial architectural and deployment overhead. We present Qwen3-VL-Seg, a parameter-efficient framework that treats the MLLM-predicted box as a semantically grounded structural prior and decodes it into pixel-level referring segmentation. At its core, a lightweight box-guided mask decoder combines multi-scale spatial feature injection, spatial-semantic query construction, box-guided high-resolution pixel fusion, and iterative mask-aware query refinement, introducing only 17M parameters (about 0.4\% of the base model). For scalable open-world training, we construct SA1B-ORS, an SA-1B-derived dataset with two subsets: SA1B-CoRS (category-oriented samples) and SA1B-DeRS (descriptive, instance-specific samples). For evaluation, we curate ORS-Bench, a manually screened benchmark with in-distribution and out-of-distribution subsets covering diverse referring expression types. Extensive experiments on referring expression segmentation, visual grounding, and ORS-Bench show that Qwen3-VL-Seg performs strongly across closed-set and open-world settings, with clear advantages on language-intensive instructions and strong out-of-distribution generalization. Evaluations on general multimodal benchmarks further show that the model broadly preserves general-purpose multimodal competence after segmentation-oriented adaptation.
Authors: Peiyu Duan, Xueqi Guo, Sepehr Farhand, Mehmet Berk Sahin, Xinyuan Zheng, James S. Duncan, Gerardo Hermosillo Valadez, Yoshihisa Shinagawa
Abstract: Accurate 3D brain MRI subtype classification benefits from both localized anatomical cues and long-range contextual reasoning. We present AGA3DNet, a report-grounded framework that incorporates brief anatomical phrases extracted from radiology reports as a soft anatomical prior channel and fuses it with a lightweight 3D CNN and multi-view xLSTM aggregation. Specifically, extracted anatomical phrases are mapped to atlas-defined regions and converted into smooth spatial priors using a signed-distance transform followed by Gaussian weighting, providing interpretable, anatomy-grounded guidance without requiring dense voxel annotations. We evaluate AGA3DNet on a retrospective institutional brain MRI cohort for abnormal subtype discrimination and compare against reproducible 3D classification baselines. AGA3DNet achieves improved overall balance across performance metrics and supports clinically interpretable localization through the prior channel. We discuss limitations related to single-cohort evaluation and the lack of large-scale public brain MRI datasets paired with radiology reports under broadly usable terms.
Authors: Zhekai Fan, Wanze Li, Jinxin Wang, Yunpeng Shi
Abstract: Translation averaging aims to recover camera locations from pairwise relative translation directions and is a fundamental component of global Structure-from-Motion pipelines. The problem is challenging because direction measurements contain no distance information, making the estimation problem highly ill-conditioned and highly sensitive to corrupted observations. In this paper, we propose TriP, a triangle-based framework for robust translation averaging. TriP first infers local relative edge scales from triangle geometry, and then synchronizes the scales of overlapping triangles in the logarithmic domain to recover globally consistent edge lengths and camera locations. By leveraging higher-order consistency across triangles, the proposed method is robust to adversarial, cycle-consistent, and other structured corruptions. In addition, TriP avoids the collapse issue without requiring any extra anti-collapse constraints, since log-scale synchronization excludes the degenerate zero-scale solution by construction. These structural advantages enable a particularly strong theory for exact location recovery. On the practical side, TriP is fully parallelizable, computationally efficient, and naturally scalable to graphs with millions of cameras. Moreover, it outperforms all previous translation averaging methods by a large margin on both synthetic and real datasets.
Authors: Laibin Chang, Shaodong Wang, Yunke Wang, Xu Zhang, Kui Jiang, Chang Xu, Bo Du
Abstract: Underwater salient object detection (USOD) plays a vital role in marine vision tasks but remains fundamentally challenging due to severe visual degradation, such as selective absorption and medium scattering. Conventional pipelines typically adopt a sequential "enhance-then-detect" paradigm. However, isolating low-level visual restoration from high-level semantic perception often leads to semantic inconsistency, where the restored images may not be optimal for detection and can even introduce task-irrelevant noise. To break this sequential bottleneck, we propose UniV2D, a Unified Vision-to-Detection Network that jointly optimizes visual restoration and salient object detection within a mutually beneficial framework. Unlike traditional methods that rely on disjointed pipelines or rigid physical priors, UniV2D introduces a semantic-driven learning paradigm: high-level saliency semantics actively guide the restoration process, while the restored visual cues reciprocally enhance saliency perception. Specifically, UniV2D features a hierarchical dual-branch architecture. It first employs a self-calibrated decoder to predict initial saliency masks alongside a mask-aware restoration module to reconstruct image content. Subsequently, a saliency-guided refinement module equipped with cross-level modulation is utilized to align structural fidelity with semantic consistency. Extensive experiments across multiple benchmarks demonstrate that UniV2D significantly outperforms state-of-the-art methods in both quantitative and qualitative evaluations, establishing a new standard for joint underwater perception.
Authors: Haoming Wang, Wei Gao
Abstract: Decades of cognitive science establish that humans navigate environments by forming cognitive maps, defined as allocentric and topology-preserving representations of 3D space. While modern Vision-Language Models (VLMs) demonstrate emergent spatial reasoning from 2D egocentric inputs, it remains unclear whether they construct an analogous 3D internal representation. In this paper, we demonstrate that current VLMs do possess a latent topological map of 3D scenes, but it is heavily overshadowed by non-geometric visual semantics, such as color and shape. By isolating this spatial subspace through cross-scene linear feature extraction, we extract a clean spatial subspace that causally controls the model's spatial outputs. We mathematically shape this latent representation and prove its correspondence to the Laplacian eigenmaps of the scene's 3D Gaussian-kernel graph, converging to the physical 3D space in the continuous limit. Motivated by this geometric identification, we further introduce a mathematically principled latent regularization method for VLMs, based on Dirichlet energy. Applying this single-term regularizer to a minimal 500-step supervised VLM fine-tuning (SFT) on simple synthetic data yields significant improvements on real-world spatial benchmarks, outperforming standard SFT and competitive baselines by up to 12.1\% in spatial tasks involving scene topology understanding. Source code is available at https://github.com/pittisl/vlm-latent-shaping
Authors: Wenbing Zhu, Jianing Liang, Linjie Cheng, Yurui Pan, Zhuhao Chen, Qingwang Yan, Yudong Cheng, Jianghui Zhang, Mingmin Chi, Bo Peng
Abstract: Industrial Anomaly Detection (IAD) is critical for quality control, but existing methods struggle with subtle, geometric defects. Standard 2D (RGB) images are sensitive to texture and lighting but often miss fine geometric anomalies. While 3D point clouds capture macro-shape, they are typically too sparse to detect micro-defects like scratches or pits. We address this fundamental data limitation by introducing Real-IAD-MVN (Multi-View Normal), a large-scale industrial dataset. By upgrading our acquisition system, Real-IAD-MVN captures high-fidelity surface normal maps from five distinct viewpoints, replacing sparse 3D data entirely. This provides a comprehensive geometric representation at a micro-detail level, making previously invisible side-wall and occluded defects explicitly detectable. Our experiments, conducted on this new dataset, first provide evidence that incorporating dense, multi-view pseudo-3D (surface normals) yields significantly better detection performance than using sparse 3D point cloud data. To further validate the dataset and provide a strong benchmark, we introduce a baseline method based on reconstruction, which learns to extract cross-modal unified prototypes from the image and normal map streams. We demonstrate that this unified prototype approach surpasses existing state-of-the-art multimodal fusion methods, highlighting the rich potential of our new dataset for advancing geometric anomaly detection.
Authors: Luqi Zhang, Zhen Dong, Bisheng Yang
Abstract: Urban spatial evolution is manifested not only through horizontal expansion but also through vertical structural changes. Consequently, jointly capturing 2D semantic changes and 3D height changes is essential for urban morphology analysis and emergency management. In practical scenarios, collecting 3D observations is often constrained by high acquisition costs and the inability to support frequent updates. The multi-temporal cross-modal input consisting of pre-event Digital Surface Model (DSM) and post-event imagery provides a practical solution for 3D change detection in high-frequency urban monitoring, disaster assessment, and emergency response scenarios. However, this setting remains challenging as imagery and DSM data exhibit significant spectral-geometric representation gaps. Moreover, modality differences may be confused with actual changes, and robust change detection requires effective fusion of semantic and geometric features from multi-temporal data. In this paper, we propose DPG-CD, a depth-prior-guided multi-temporal cross-modal fusion framework for joint 2D semantic and 3D height change detection. Specifically, an estimated depth prior is introduced into the imagery to mitigate the modality gap with DSM. A gated fusion mechanism then selectively injects geometric cues from depth prior while preserving discriminative spectral representations. Subsequently, a multi-stage cross-temporal cross-modal feature fusion architecture is employed to extract change-aware features. Finally, a multi-task decoder jointly predicts 2D semantic changes and 3D height changes, complemented by an auxiliary DSM prediction task to improve structural consistency and height estimation accuracy. Experiments on two public datasets, Hi-BCD and 3DCD, and a new dataset, NYC-MMCD, demonstrate that DPG-CD outperforms state-of-the-art methods on both 2D and 3D change detection tasks.
Authors: Yuchen He, Jing Zhang
Abstract: Referring Audio-Visual Segmentation (Ref-AVS) seeks to localize and segment target objects in video frames based on visual, auditory, and textual referring cues. The task is challenging because the relevance of different modalities varies across referring expressions and scenes, while existing methods typically treat multimodal cues as homogeneous inputs for fusion, prompting, or reasoning, making them vulnerable to irrelevant or misleading modalities. To address this problem, we propose PRIMED, inspired by the biased competition theory in cognitive neuroscience, which explicitly models both visual perception and language-driven prior modulation, and enables more accurate Ref-AVS by adaptive modality suppression. Specifically, a Modality Prior Decoder first estimates whether the referring expression relies primarily on audio, vision, or their joint interaction, generating a modality prior to adaptively guide high-level attention. A Token Distiller further extracts compact global visual tokens from high-level features and shares them across Competition-aware Cross-modal Fusion modules to provide hierarchical global context. Additionally, we introduce a Spatial-Aware Semantic Alignment loss to further enhance foreground-background discrimination through contrastive learning. Extensive experiments on the Ref-AVS benchmark demonstrate that PRIMED achieves state-of-the-art overall performance.
Authors: Han Jang, Junhyeok Lee, Heeseong Eum, Joon Jang, Yoseob Han, Seung Hong Choi, Kyu Sung Choi
Abstract: Precise molecular subtyping of gliomas, including isocitrate dehydrogenase (IDH) mutation and 1p/19q codeletion, directly guides surgical and therapeutic decisions, yet currently relies on invasive tissue sampling. Deep learning on structural MRI has emerged as a non-invasive alternative, but anatomy-only approaches cannot capture the hemodynamic signatures that distinguish molecular subtypes. Radiogenomics based on dynamic susceptibility contrast (DSC) MRI holds immense potential for non-invasively characterizing glioma molecular subtypes, yet clinical deployment has been hindered by inter-site variability and the limitations of voxel-wise analysis. We introduce HiPerfGNN, a framework that first learns discrete hemodynamic representations from raw time-intensity curves using a vector-quantized variational autoencoder (VQ-VAE). These quantized perfusion codes define coarse-level graph nodes representing functional tumor habitats, each of which is hierarchically subdivided into fine-level subregions guided by structural MRI. A hierarchical graph neural network then propagates information across scales for molecular prediction. On an internal cohort (n=475), the model achieved AUCs of 0.96 (IDH), 0.89 (1p/19q), and 0.84 (WHO grade), and maintained robust IDH performance (AUC 0.89) on an independent external cohort (n=397) without recalibration. Gradient-based saliency analysis confirms biologically grounded attention patterns aligned with known glioma pathophysiology. Our results demonstrate the added value of integrating perfusion dynamics into radiogenomic pipelines for glioma molecular subtyping. Code is available at https://github.com/janghana/HiPerfGNN.
Authors: Kai Zheng, Hang-Cheng Dong, Jiatong Pan, Zhenkai Wu, Fupeng Wei, Wei Zhang
Abstract: Remote sensing change detection is pivotal for urban monitoring, disaster assessment, and environmental resource management. Yet, unimodal deep learning methods frequently confuse genuine semantic changes with visually similar but irrelevant variations. Recent multimodal approaches incorporate text as auxiliary supervision, but their descriptions are either semantically coarse and unstructured or model-generated and thus noisy. Critically, all of them overlook a simple fact: fine-grained change semantics are already implicitly encoded in the ground-truth mask labels that come standard with every change detection dataset. These masks know where the change happened, what the land-cover types were before and after, how the transition occurred, and how many objects were involved. In this paper, we propose S2M, a framework that obtains structured textual features directly from change labels at zero additional annotation cost. Specifically, each change region is automatically transcribed into a semantic quadruple (where, what, how, how many) and converted into several fixed-template text descriptions, providing precise, dense, and noise-free multimodal supervision. We adopts a two-stage training strategy to fine-tune on remote sensing imagery firstly for robust domain-specific representation, after which a multimodal decoder with a bi-directional contrastive loss is introduced to achieve deep alignment between visual features and structured textual embeddings. To validate our method, we construct Gaza-Change-v2, a new multi-class change detection (MCD) dataset about the Gaza Strip. On this MCD dataset, S2M achieves a Sek of 17.80\% and an F$_{\text{scd}}$ of 66.14\%, notably surpassing even multimodal methods that leverage large language models. Our work demonstrates that masks can indeed talk. They tell us exactly what, where, how, and how many changes have occurred.
Authors: Min Chen, Wei Guo, Bin Wang, Wen Li, Tong Fang, Jinbo Zhang, Junqi Zhao, Hong Kuang, Han Hu, Xuming Ge, Qing Zhu, Bo Xu
Abstract: Sparse-view satellite image surface reconstruction remains highly challenging, fundamentally because the reliability of multi-view matching under satellite imaging conditions is strongly spatially heterogeneous. Affected by large photometric differences, weak textures, and repetitive textures, multi-view geometric constraints are often sparse, unevenly distributed, and locally unreliable. Although 2D Gaussian Splatting (2DGS) is more suitable than 3D Gaussian Splatting (3DGS) for the explicit representation of continuous surfaces, research on generalizable feed-forward 2DGS frameworks for sparse-view satellite surface reconstruction is still lacking. To address this issue, we propose SatSurfGS, a generalizable sparse-view surface reconstruction method for satellite imagery based on 2DGS. The proposed method builds a coarse-to-fine Gaussian attribute prediction framework and explicitly models local geometric reliability at three levels: feature learning, Gaussian parameter estimation, and training optimization. Specifically, we propose a confidence-aware monocular multi-view feature fusion module to adaptively integrate monocular priors and multi-view matching features according to local confidence; a cross-stage self-consistency residual guidance module to stabilize stage-wise Gaussian parameter refinement using the residual between the rendered height map from the previous stage and the current-stage MVS height map, together with confidence information; and a confidence bidirectional routing loss to achieve differentiated allocation of geometric and appearance supervision. Experiments on satellite datasets show that the proposed method achieves improved rendering quality, surface reconstruction accuracy, cross-dataset generalization, and inference efficiency compared with representative generalizable baselines and competitive per-scene optimization methods.
Authors: Fuxin Duan, Hui Wang
Abstract: We present PicoEyes, a unified gaze estimation framework that directly predicts all key attributes of gaze, including 3D eye parameters, eye-region segmentation, optical axis, visual axis, and depth maps, from either monocular or binocular inputs. The framework simultaneously addresses calibration, gaze forecasting, and varying device postures, while also supporting 3D eye reconstruction via joint estimation of eye parameters and depth maps in an end-to-end manner. In addition, we introduce a large-scale multi-view near-eye dataset containing comprehensive 2D and 3D annotations under diverse conditions, including train, test, rewear-test, and calibration sessions. Extensive experiments demonstrate that PicoEyes achieves state-ofthe-art performance, consistently outperforming both academic and industrial gaze tracking methods across nocalibration, calibration, rewear-after-calibration, and forecasting settings. This work establishes a practical, end-toend paradigm for robust and generalizable gaze estimation in mixed reality (MR) applications.
Authors: Huaiyuan Qin, Muli Yang, Gabriel James Goenawan, Peng Hu, Chen Gong, Xi Peng, Hongyuan Zhu
Abstract: A recent work shows that Attention Transfer, which transfers only the attention patterns from a pre-trained teacher Vision Transformer (ViT) to a randomly initialized standard student ViT, is sufficient to recover the full benefit of the teacher's pre-trained weights. We revisit this finding on a comprehensive benchmark of 20 teachers from 11 well-known ViT families and reveal that Attention Transfer is not universally effective. While 7 families transfer successfully, 4 consistently fail, falling up to 5.1\% below the from-scratch no-transfer baseline. Further results demonstrate that this failure is family-consistent across model sizes, and persists under extended training durations, different transfer datasets, and out-of-distribution evaluations. Controlled analyses then consistently localize the problem to the attention-routing channel, indicating that the key issue is not whether the student can match the teacher's attention patterns, but whether the matched patterns remain functional for the student. Crucially, we identify architectural mismatch between the pre-trained teacher and the standard student as the primary mechanism. By adding only the teacher's native architectural components to the student in a randomly initialized state, we completely reverse the failure for all 4 families. Notably, these components alone do not improve from-scratch training, confirming that they specifically unlock the usability of the teacher's attention. We further systematically show that this failure is not explained by the inadequate choice of transfer loss or by differences in pre-training recipes. Our findings refine the prevailing understanding of attention in ViT representations: attention is sufficient \textit{only} when the student architecture matches the teacher.
Authors: Jun Dai, Renbiao Jin, Bo Xu, Yutian Chen, Linning Xu, Mulin Yu, Tianfan Xue, Shi Guo
Abstract: 3D reconstruction methods such as 3D Gaussian Splatting (3DGS) and Neural Radiance Fields (NeRF) achieve impressive photorealism but fail when input images suffer from severe motion blur. While event cameras provide high-temporal-resolution motion cues, existing event-assisted approaches rely on low-resolution sensors and strict synchronization, limiting their practicality for handheld 3D capture on common devices, such as smartphones. We introduce a flexible, high-resolution asynchronous RGB-Event dual-camera system and a corresponding reconstruction framework. Our approach first reconstructs sharp images from the event data and then employs a cross-domain pose estimation module based on the Visual Geometry Transformer (VGGT) to obtain robust initialization for 3DGS. During optimization, we employ a structure-driven event loss and view-specific consistency regularizers to mitigate the ill-posed behavior of traditional event losses and deblurring losses, ensuring both stable and high-fidelity reconstruction. We further contribute AsyncEv-Deblur, a new high-resolution RGB-Event dataset captured with our asynchronous system. Experiments demonstrate that our method achieves state-of-the-art performance on both our challenging dataset and existing benchmarks, substantially improving reconstruction robustness under severe motion blur. Project page: https://openimaginglab.github.io/AsyncEvGS/
Authors: Bincheng Peng, Guang Li, Ping Liu, Takahiro Ogawa, Miki Haseyama
Abstract: Dataset distillation compresses a large training set into a small synthetic set that preserves downstream training utility. While most existing methods target training networks from scratch, modern visual transfer learning often uses frozen pre-trained encoders followed by lightweight linear probing. Existing distillation methods for this setting either unroll iterative linear-probe updates with trajectory-based gradient matching, or rely on closed-form formulations originally designed for from-scratch training with neural-tangent-kernel (NTK) approximations. Neither route exploits the fact that frozen-feature linear probing admits a closed-form solution determined directly by the pre-trained features themselves, with no infinite-width approximation and no inner-loop trajectory. We propose Closed-Form Linear-Probe Dataset Distillation (CLP-DD), a bilevel formulation that computes the linear probe induced by the synthetic set with a sample-space kernel ridge solver. The synthetic images are then updated by evaluating this induced classifier on real features through a temperature-scaled softmax cross-entropy, where the classifier columns act as learned class anchors in feature space. We further show that the choice of outer objective is decisive: pairing the closed-form inner solver with a standard MSE outer loss substantially underperforms trajectory-based methods, while the discriminative outer loss closes most of the gap. On ImageNet-100 with four pre-trained backbones, CLP-DD substantially improves over LGM without DSA and approaches LGM with DSA at a fraction of the computational cost. On ImageNet-1K, CLP-DD matches or surpasses LGM with DSA on three of four backbones while running roughly $14\times$ faster and using less than one-eighth of the GPU memory.
Authors: Bozhou Zhang, Nan Song, Yuang Wang, Jiankang Deng, Xiatian Zhu, Li Zhang
Abstract: Current end-to-end autonomous driving planners are fundamentally reactive: they condition on historical and present observations to predict future actions. We argue that autonomous agents should instead imagine future scenes before deciding, just as human drivers mentally simulate ``what will happen next" before acting. We introduce ForeSight, a foundation world model centric planning framework that reframes autonomous driving as anticipatory decision-making. Rather than treating world models as auxiliary components, ForeSight makes future scene imagination the primary driver of action prediction. Our approach operates in two stages: (1) generating plausible future visual worlds via a pretrained world model, and (2) planning actions conditioned on these imagined futures. This paradigm shift from ``what should I do now?" to ``what will happen, and how should I respond?" enables genuinely anticipatory rather than reactive planning. By grounding decisions in anticipated contexts rather than present observations alone, ForeSight navigates dynamic, interactive scenarios more effectively. Extensive experiments on NAVSIM and nuScenes demonstrate that explicit future imagination significantly outperforms previous state-of-the-art alternatives, validating our foresight-driven approach.
Authors: Chamuditha Jayanga Galappaththige, Jason Lai, Timothy Patten, Donald Dansereau, Niko Suenderhauf, Dimity Miller
Abstract: Scene change detection methods built on Gaussian splatting universally follow a render-then-compare paradigm: the pre-change scene is rendered into 2D and compared against post-change images via pixel or feature residuals. This change detection problem with Gaussian Splatting has been treated as a question about pixels; we treat it as a question about primitives. We provide direct evidence that native primitive attributes alone -- position, anisotropic covariance, and color -- carry sufficient signal for scene change detection. What makes primitive-space comparison hard is the under-constrained nature of Gaussian splatting representation: independent optimizations yield primitive solutions whose count, positions, shapes, and colors differ even where nothing has changed. We address this challenge with anisotropic models of geometric and photometric drift, complemented by a per-primitive observability term that reflects the extent to which each Gaussian is constrained by the camera geometry. Operating directly on primitives gives our method, GD-DIFF, two properties that distinguish it from render-then-compare methods. First, change maps are multi-view consistent by construction, where prior work had to learn this through an additional optimization objective. Second, geometric and appearance changes are scored separately, identifying not just where but what kind of change occurred, distinguishing structural changes (e.g., an added object) from surface-level ones (e.g., a color change) without supervision or external model dependencies. On real-world benchmarks, GS-DIFF surpasses the prior state-of-the-art approach by approximatelt 17% in mean Intersection over Union.
Authors: Qianwen Ma, Yang Xu, Shangwei Deng, Xiaobo Li, Haofeng Hu
Abstract: Infrared small target detection (IRSTD) remains challenging due to the scarcity of useful target cues and the presence of severe background clutter. Most current methods rely on conventional feature learning and local interaction modeling, where features are represented in Euclidean space. However, such designs may still be limited in describing the subtle differences of weak targets and the contextual relations between targets and backgrounds. To address these limitations, we propose LoHGNet, an IRSTD network that integrates Lorentz geometric encoding with high-order relation learning. By introducing Lorentz manifold based feature learning, LoHGNet offers a different feature representation from conventional IRSTD methods and provides new discriminative cues for IRSTD. Specifically, a Lorentz encoding branch is constructed with the Geometric Attention Guided Lorentz Residual Convolution Module (GA-LRCM) to perform feature modeling under hyperbolic geometric constraints and enhance the hierarchical geometric representation capability of weak targets. Subsequently, the hyperbolic features are mapped into the Euclidean tangent space through logarithmic mapping, and a High-Order Relation Learning Module (HORL) is designed to model the high-order contextual dependencies between targets and backgrounds via hypergraph construction, thereby improving target discrimination in complex backgrounds. Experimental results on three datasets demonstrate that the proposed LoHGNet achieves competitive performance in both detection accuracy and adaptability to complex scenes. The code will be available at https://github.com/Kingwin97.
Authors: Wei Jiang, Feng Liu, Nan Ye, Hongfu Sun
Abstract: Adapting foundation models to medical segmentation typically requires either backbone fine-tuning or high-capacity task-specific decoders, both of which are difficult to fit reliably when annotations are scarce. We show that frozen DINOv3 features already contain useful structural and boundary cues for medical segmentation, and that the main bottleneck lies in how these features are read out. We propose DINO-MVR, a Multi-View Readout framework for annotation-efficient medical segmentation. DINO-MVR trains only lightweight MLP probes on features from the final three transformer blocks of a frozen DINOv3 backbone, without updating the backbone itself. At inference, each input is interpreted through complementary resolutions and test-time augmentations, whose probability maps are combined by entropy-weighted fusion and refined with simple spatial regularization. For volumetric inputs, Gaussian z-axis smoothing further improves inter-slice consistency. Under fixed evaluation protocols on endoscopy, dermoscopy, and MRI benchmarks, DINO-MVR achieves strong readout-only performance, including 0.895 Dice on Kvasir-SEG, 0.897 Dice on ISIC 2018, and 0.908 Dice on BraTS FLAIR whole-tumor segmentation. With only five annotated BraTS patients, it recovers 98.4% of the performance obtained by the 40-patient BraTS reference run. These results suggest that frozen self-supervised vision backbones can support accurate medical segmentation when paired with an effective multi-view readout.
Authors: Selin Yildirim, Subhajit Dutta Chowdhury, Mohammad Mahdi Kamani, Vikram Appia, Deming Chen
Abstract: Autoregressive generation is a powerful approach for high-fidelity image synthesis, but it remains computationally demanding and slow even on the most advanced accelerators. While speculative decoding has been explored to mitigate this bottleneck, existing approaches fail to achieve efficiency gains comparable to those observed in text generation. A key limitation is the target model's high uncertainty during image generation, which leads to high draft token rejection rates. In this work, we identify previously overlooked patterns in the target model's behavior that emerge naturally in tree-based speculative decoding. Specifically, we formalize two properties, semantic interchangeability and convergence, arising from the redundancies in the target model's hidden state representations. By capturing these redundancies across the depth and breadth of the predicted token tree, our method identifies principled opportunities for acceptance relaxation without requiring additional training. Additionally, we enhance standalone drafter performance by injecting the redundancy signals from the target model into drafter training with minimal modification. We evaluate our approach across multiple text-to-image models and drafter architectures. Results show that CASCADE achieves state-of-the-art speedups for drafter-based speculative decoding, with up to 3.6x acceleration, while maintaining image quality and text-prompt fidelity.
Authors: Dat Le, Khoa Nguyen, Xin Wang, Shu Hu
Abstract: Recent advances in generative AI have democratized video creation at scale. AI-generated videos, including partially manipulated clips across visual and audio channels, pose escalating risks of semantic distortion and misuse, which motivates the need for reliable detection tools. Most existing AI-generated video detectors remain limited by single- or partial-modality of data modeling and the lack of fine-grained temporal forgery localization. To address these challenges, our primary novelty introduces a core architecture that jointly integrates an LMM semantic branch with a spatio-temporal (ST) visual branch and a multi-scale partial-spoof (PS) audio branch. This multi-modal approach enables simultaneous detection and fine-grained temporal localization of partially manipulated AI-generated video forgeries. Extensive experiments show that this approach outperforms existing state-of-the-art methods.
Authors: Zhixue Song, Boyan Han, Yiwei Wang, Chi Zhang
Abstract: Recent advancements in visual context compression enable MLLMs to process ultra-long contexts efficiently by rendering text into images. However, we identify a critical vulnerability inherent to this paradigm: lowering image resolution inadvertently catalyzes jailbreaking. Our experiments reveal that the safety defenses of SOTA models deteriorate sharply as resolution degrades, surprisingly persisting even when text remains legible. We attribute this to ``Cognitive Overload'', hypothesizing that the effort required to decipher degraded inputs diverts attentional resources from safety auditing. This phenomenon is consistent across various visual perturbations, including noise and geometric distortion. To address this, we propose a simple ``Structured Cognitive Offloading'' strategy that mitigates these risks by enforcing a serialized pipeline to decouple visual transcription from safety assessment. Our work exposes a significant risk in vision-based compression and provides critical insights for the secure design of future MLLMs.
Authors: Haewon Jeon, Si-Hyeon Lee
Abstract: Distilled diffusion models accelerate image generation by reducing the number of denoising steps, but often suffer from degraded image quality. To mitigate this trade-off, test-time optimization methods improve quality, yet their iterative nature incurs substantial computational overhead and leads to slow inference, limiting practical usability. Recent hypernetwork-based approaches amortize this process during training, but still require costly noise modulation in high-dimensional latent spaces. In this work, we propose LENS (Low-frequency Eigen Noise Shaping), an efficient noise modulation framework that operates in a low-dimensional subspace. Our approach is motivated by the observation that low-frequency components of the noise largely determine the global structure and visual fidelity of generated images. Based on this observation, we provide a theoretical justification for restricting modulation to the low-frequency subspace and derive a principled training objective. Building on this, LENS employs a lightweight, standalone network to selectively modulate these components, enabling efficient and targeted noise modulation. Extensive experiments demonstrate that LENS achieves competitive image quality while reducing FLOPs by 400-700$\times$, model parameters by 25-75$\times$, and inference-time overhead by 10-20$\times$ compared to prior methods.
Authors: Nandhana Sunil, Abhirami R Iyer, Avirup Mandal
Abstract: Multi-view mesh reconstruction remains a core challenge in computer graphics and vision, especially for recovering high-frequency geometry from sparse observations. Recent methods such as 3D Gaussian Splatting (3DGS) and Neural Radiance Fields (NeRF) rely on post-processing for mesh extraction, thereby limiting joint optimization of geometry and appearance. Implicit Moving Least Squares (IMLS) instead enables direct conversion of point clouds into signed distance and texture fields, supporting end-to-end reconstruction and rendering. However, existing IMLS formulations use exponential kernels that struggle with high-frequency detail. We introduce a compact polynomial kernel with local support and greater flexibility, allowing better control over frequency content and improved geometric fidelity. To further enhance fine details, we incorporate stochastic regularization with Laplacian filtering. Together, these improve the preservation of high-frequency structure while maintaining stable optimization. Experiments show state-of-the-art performance in both surface reconstruction and rendering, yielding more accurate geometry and sharper visuals from multi-view data.
Authors: Jeimin Jeon, Hyunju Lee, Bumsub Ham
Abstract: Transformer architecture search (TAS) discovers optimal vision transformer (ViT) architectures automatically, reducing human effort to manually design ViTs. However, existing TAS methods suffer from the feature collapse problem, where subnets within a supernet fail to learn subnet-specific features, mainly due to the shared weights in a supernet, limiting the performance of individual subnets. To address this, we propose TAS-LoRA, a novel method that introduces parameter-efficient low-rank adaptation (LoRA) to enable subnet-specific feature learning, while maintaining computational efficiency. TAS-LoRA incorporates a Mixture-of-LoRAExperts (MoLE) strategy, where a lightweight router dynamically assigns LoRA experts based on subnet architectures, and introduces a group-wise router initialization technique to encourage diverse feature learning across experts early in training. Extensive experiments on ImageNet and several transfer learning benchmarks, including CIFAR-10/100, Flowers, CARS, and INAT-19, demonstrate that TAS-LoRA mitigates feature collapse effectively, improving performance over state-of-the-art TAS methods significantly.
Authors: Van-Anh Nguyen, Anh Tuan Bui, Tamas Abraham, Junae Kim, Amardeep Kaur, Rollin Omari, Thuy-Trang Vu, Dinh Phung
Abstract: Generative personalization often suffers from the semantic collapsing problem (SCP), where a learned personalized concept overpowers the rest of the text prompt, causing the model to ignore important contextual details. To address this, we first analyze the underlying cause, revealing that the semantic drift responsible for SCP is not random but is concentrated within a specific low-dimensional subspace. We also discover that the personalization process perturbs the embedding of the original base concept, making it an unstable reference point. Based on these insights, we introduce Test-time Embedding Adjustment with Adaptive Subspace Projection (AdaptSP), a training-free method that uses the stable, pre-trained embedding as an anchor. AdaptSP isolates the semantic drift and projects it onto the identified subspace, performing a precise adjustment that mitigates SCP while maintaining the subject identity. Our experiments show that this targeted approach significantly improves prompt fidelity and contextual alignment.
Authors: Qiaoyi Yang, Chaoyi Zhou, Xi Liu, Run Wang, Minghui Xu, Mert D. Pes\'e, Feng Luo, Yuhao Xu, Zhi-Qi Cheng, Qiushi Chen, Hairong Qi, Siyu Huang
Abstract: Accurate Digital Surface Model (DSM) reconstruction from satellite imagery is critical for applications such as disaster response, urban planning, and large-scale geographic mapping. Existing approaches face a fundamental trade-off: optimization-based methods achieve strong accuracy but require hours of per-scene computation, while generalizable geometry foundation models offer near-instant inference but fail to generalize to satellite imagery due to the domain gap introduced by the Rational Polynomial Camera (RPC) model and mismatched depth scale distributions. We present Sat3R, a feed-forward framework that bridges this gap via RPC-aware metric depth fine-tuning of Depth Anything V2 using the Scale-Invariant Logarithmic (SiLog) loss. By constructing physically consistent pseudo depth supervision from RPC geometry, Sat3R adapts a monocular depth foundation model to the satellite domain without per-scene optimization. Experiments on the DFC2019 benchmark demonstrate that Sat3R reduces MAE by 38% over zero-shot feed-forward baselines and achieves competitive accuracy against optimization-based methods, while delivering over 300x speedup. Sat3R demonstrates that feed-forward models, when properly adapted to the satellite domain, can match optimization-based accuracy at a fraction of the computational cost, paving the way for practical large-scale satellite DSM reconstruction.
Authors: Jiaju Han, Chao Li, Chengyin Hu, Qike Zhang, Xuemeng Sun, Xin Wang, Fengyu Zhang, Xiang Chen, Yiwei Wei, Jiahuan Long, Jiujiang Guo
Abstract: Multimodal RAG systems increasingly rely on vision-language retrievers to ground visual queries in external textual evidence. Existing adversarial studies on RAG mainly manipulate the retrieval corpus or memory, while attacks on vision-language and remote sensing models typically target end-task predictions. Input-space threats to the evidence retrieval stage of remote sensing multimodal RAG remain underexplored. To address this gap, we introduce CloudWeb, an atmospheric retrieval hijacking attack that modifies only the input image while keeping the retriever, generator, and knowledge base fixed at deployment. CloudWeb overlays parameterized cloud- and haze-like patterns on remote sensing images and optimizes them with a retrieval-oriented objective that pulls adversarial image embeddings toward target atmospheric evidence, suppresses source-scene evidence, enforces rank separation, and regularizes naturalness and coverage. To the best of our knowledge, this is the first study of retrieval-stage atmospheric evidence hijacking in remote sensing multimodal RAG. We evaluate CloudWeb on a seven-dataset remote sensing RAG benchmark with five CLIP-style retrievers, including GeoRSCLIP, RemoteCLIP, OpenAI CLIP, and OpenCLIP, together with downstream vision-language generators. Across retrievers, CloudWeb consistently outperforms clean retrieval, handcrafted atmospheric baselines, random cloud perturbations, and fixed variants in injecting weather-related evidence into top-ranked results. On GeoRSCLIP ViT-B/32, Weather@5 increases from 0.71\% to 43.29\%. Downstream generation further shows measurable weather hallucination and semantic shift, indicating that retrieval-stage hijacking can propagate to the final RAG response. These findings reveal a practical failure mode: natural-looking atmospheric changes can compromise evidence retrieval before generation begins.
Authors: Yecong Wan, Fan Li, Mingwen Shao, Wangmeng Zuo
Abstract: Generalizable novel view synthesis aims to render unseen views from uncalibrated input images without requiring per-scene optimization. Recent feed-forward approaches based on 3D Gaussian Splatting have achieved promising efficiency and rendering quality. However, most of them assign a fixed number of Gaussians to each pixel or voxel, ignoring the spatially varying complexity of real-world scenes. Such uniform allocation often wastes Gaussian primitives in smooth regions while providing insufficient capacity for fine structures, complex geometry, and high-frequency details. This motivates us to predict region-dependent primitive cardinalities rather than impose a fixed primitive budget everywhere, enabling a more expressive yet compact 3D scene representation. Therefore, we propose SplatWeaver, a generalizable novel view synthesis framework that is able to dynamically allocate Gaussian primitives over different regions in a feed-forward manner. Specifically, SplatWeaver introduces cardinality Gaussian experts and a pixel-level routing scheme, wherein each expert specializes in producing a specific number of primitives from 0 to M, and the routing scheme coordinates these experts to adaptively determine how many Gaussian primitives should be allocated to each spatial location. Moreover, SplatWeaver incorporates a high-frequency prior with attendant guidance module and routing regularization to stabilize expert selection and promote complexity-aware allocation. By leveraging high-frequency structural cues, the routing process is encouraged to assign more Gaussian primitives to fine structures, complex geometry, and textured regions, while suppressing redundant primitives in smooth areas. Extensive experiments across diverse scenarios show that SplatWeaver consistently outperforms state-of-the-art methods, delivering more faithful novel-view renderings with fewer Gaussian primitives.
Authors: Jiaxuan Gao, Yongjian Guo, Zhong Guan, Wen Huang, Wanlun Ma, Xi Xiao, Junwu Xiong, Sheng Wen
Abstract: The integration of Vision-Language-Action (VLA) models with World Models has gained increasing attention. One representative approach treats learned World Models as generative simulators, enabling policy optimization entirely within "imagination." However, when deployed as simulators for specific environments such as the LIBERO benchmark, existing World Models often suffer from poor generalization and long-horizon error accumulation. During closed-loop rollouts, these models are highly sensitive to initial-state perturbations; minor changes in color, illumination, and other visual factors can trigger cascading hallucinations, leading to severe blurriness or overexposure. Moreover, long-horizon error accumulation further degrades the quality and fidelity of predicted future states. These issues limit the reliability of World Models as simulators. To mitigate these problems, we propose Sword, a robust World Model framework. Our method introduces Structure-Guided Style Augmentation to disentangle the visual textures of interactive environments from task-relevant dynamics, thereby improving generalization. We further propose Dynamic Latent Bootstrapping, which maintains consistency between training and inference while keeping memory consumption low. Extensive experiments on the LIBERO benchmark show that our method significantly outperforms the baseline WoVR in terms of generalization, generation quality, robustness, fidelity, and the success rate of reinforcement-learning post-training for VLA models.
Authors: Dongchuan Ran, Linyu Ou, Xueheng Li, Wenwen Tong, Chenxu Guo, Hewei Guo, Kaibing Wang, Lewei Lu
Abstract: Existing Multimodal Large Language Models (MLLMs) remain primarily reactive, failing to continuously perceive environments or proactively assist users. While emerging benchmarks address proactivity, they are largely confined to alert scenarios, neglect personalized context, and fail to evaluate the precise timing of human-machine interactions (HMI).In this paper, we introduce EgoPro-Bench, a novel benchmark for training and evaluating proactive interaction capabilities based on streaming egocentric videos; it comprises 2,400 videos in the evaluation set and over 12,000 videos in the training set.Unlike previous works, EgoPro-Bench leverages simulated user profiles to generate diverse user intentions and to construct high-fidelity HMI data across 12 distinct domains.Subsequently, we propose a specialized evaluation protocol and metrics, train proactive interaction models designed for efficient reasoning and low-latency interaction on streaming video data, and conduct comprehensive evaluations.Furthermore, we introduce an interaction principle termed "short thinking, better interaction", which allocates a limited token budget prior to intent recognition, thereby enhancing interaction performance.The experiments demonstrate that EgoPro-Bench substantially enhances the intention understanding capabilities of MLLMs and enables accurate identification of appropriate timings for HMI, thereby laying a solid foundation for next-generation user-centric proactive interactive agents.
Authors: Rui Fang, Hsi-Wen Chen, Ming-Syan Chen
Abstract: Vision Transformers (ViTs) achieve strong performance across vision tasks, yet their deployment with low-precision early exiting remains fragile. Existing quantization methods assume static full-depth execution, making them unstable when exit decisions are perturbed by quantization noise, which can amplify errors along dynamic inference paths. In this paper, we introduce Amortized-Precision Quantization (APQ), a utilization-aware formulation that accounts for layer-wise stochastic exposure to quantization noise and reveals depth-precision trade-offs. Building on APQ, we propose Mutual Adaptive Quantization with Early Exiting (MAQEE), a bi-level framework that jointly optimizes exit thresholds and bit-widths under explicit risk control to improve inference stability. MAQEE establishes a superior Pareto frontier in the accuracy-efficiency trade-off, reducing BOPs by up to 95% while maintaining accuracy and outperforming strong baselines by up to 20\% across classification, detection, and segmentation tasks.
Authors: Yang Wu, Zhaojiang Liu, Qiang Meng, Youquan Liu, Renliang Weng, Jianjun Qian, Jian Yang, Jin Xie
Abstract: World models, which simulate environmental dynamics and generate sensor observations, are gaining increasing attention in autonomous driving. However, progress in LiDAR-based world models has lagged behind those built on camera videos or occupancy data, primarily due to two core challenges: the inherent disorder of LiDAR point clouds and the difficulty of distinguishing dynamic objects from static structures. To address these issues, we propose GEM: a Generative LiDAR world model that leverages deformable mamba architecture, significantly improving fidelity and imaginative capability. Specifically, leveraging the structural similarity between sequential laser scanning and Mamba's processing mechanism, we first tokenize LiDAR sweeps into compact representations via a custom LiDAR scene tokenizer. After unsupervised disentanglement of tokenized features via a dynamic-static separator, a tri-path deformable Mamba is introduced to perform selective scanning and adaptive gating fusion over the disentangled features, leading to enhanced spatial-temporal understanding of the world evolution. Optionally, a planner and a BEV layout controller can be integrated to explore the model's capability for autonomous rollout and its potential to generate ``what-if" scenarios. Extensive experiments show that GEM achieves state-of-the-art performances across diverse benchmarks and evaluation settings, demonstrating its superiority and effectiveness. Project page: https://github.com/wuyang98/GEM.
Authors: Yuan Zhang, Chenyi Li, Guoqing Ma, Jiajun Zha, Yuanming Yang, Bo Wang, Wei Tang, Wenbo Li, Haoyang Huang, Nan Duan
Abstract: Sampling from pretrained diffusion and flow-matching models typically requires many forward passes to generate diverse and high-fidelity images. Existing distillation methods often rely on multiple auxiliary networks, carefully designed training stages, or complex optimization pipelines. In this work, we revisit the recently proposed Drifting Model objective and show that a single drifting loss can be directly used to simplify one step distillation. A key observation is that the pretrained diffusion teacher itself already provides a strong representation space. Unlike the original Drifting Model, which relies on an additional pretrained feature extractor, we use intermediate hidden states of the pretrained teacher model as the feature representation. This removes the need for training or introducing an extra representation network while preserving a semantically meaningful feature geometry for drifting. Furthermore, we introduce a lightweight mode coverage loss to mitigate mode collapse during distillation and encourage the student generator to cover diverse teacher-supported regions. Extensive experiments on ImageNet and SDXL demonstrate that our method achieves efficient one step generation with competitive image quality and diversity, achieving FID scores of 1.58 on ImageNet-64$\times$64 and 18.4 on SDXL, while substantially simplifying the overall distillation framework.
Authors: Wei Huang, Joyce Huang
Abstract: We introduce GC-ART (Global Curve Adaptive Rational Tone-mapping), a lightweight differentiable pre-processing module for robust image classification. GC-ART predicts an endpoint-pinned rational tone curve from per-channel soft histograms using a 643-parameter MLP, then applies the curve pointwise before the classifier. The module is trained end-to-end with cross-entropy and a soft monotonicity penalty. On CIFAR-10 with a CIFAR-style ResNet-18, GC-ART matches clean accuracy with the unenhanced baseline and other learned enhancers, improves over the baseline on multiplicative darkening, and achieves the best learned-method result on contrast corruption (48.45% vs. 46.27% for the baseline and 47.13% for Zero-DCE++). These results suggest that histogram-conditioned rational curves can learn useful global tone corrections, including contrast-expanding behavior, while preserving edge locations by construction through pointwise mapping. GC-ART also uses substantially fewer FLOPs than convolutional learned enhancers at 32 x 32. The current hyperparameters are untuned, leaving room for systematic improvement.
Authors: Junwei Wen, Deshui Miao, Guangming Lu, Xin Li, Wenjie Pei
Abstract: Video Reasoning Segmentation (VRS) aims to segment target objects in videos based on implicit instructions that convey human intent and temporal logic. Existing MLLM-based methods predict masks with a [SEG] token after selecting frames via simple sampling or an auxiliary MLLM, where limited supervision and frame-language similarity rules often yield narrow-scope keyframe choices that weaken holistic temporal understanding and lead to brittle localization in complex multi-object scenes. To address these issues, we introduce RCoT-Seg, a video-of-thought framework that factorizes VRS into temporal video reasoning (TVR) and keyframe target perception (KTP), explicitly separating temporal reasoning from spatial perception. Specifically, in the TVR stage, an agentic keyframe selection module, initialized with a curated CoT-start corpus and refined by GRPO under task-aligned rewards, is proposed to generate and reselect the keyframe through self-evaluation, strengthening moment localization and temporal reasoning. In the KTP stage, RCoT-Seg performs high-resolution segmentation on the selected frame and propagates masks with SAM2-based methods across the sequence, replacing heuristic sampling and external selectors while improving spatial precision and inter-frame consistency. Extensive experimental results demonstrate that the proposed RCoT-Seg achieves favorable performance against the state-of-the-art methods. The code and models will be publicly released at https://github.com/Victor-wjw/RCoT-Seg.
Authors: Ziheng Zhou, Yang Wang, Nan Wang, Chengliang Wu, Jun Yan
Abstract: The decline of global shellfish biodiversity poses a severe threat to coastal ecosystems. Although artificial intelligence (AI) technologies show potential for automated ecological monitoring, existing marine benthic datasets often lack adaptation to the complexities of real underwater environments (e.g., variable lighting conditions and diverse species postures), posing challenges for the robust generalization of vision models in practical ecological monitoring. To address this problem, we construct ShellfishNet, a comprehensive image benchmark dataset designed specifically for real-world ecological monitoring constraints. Comprising 8,691 images across 32 taxa, this dataset includes a curated subset annotated with descriptive captions. It is constructed through field photography and web scraping, encompassing samples from complex real-world environments. Based on this benchmark, we systematically evaluate 80 representative neural network models, including Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), State Space Models (SSMs), and Self-Supervised Learning (SSL) methods. Furthermore, we evaluate the performance of fine-grained visual categorization (FGVC) models and investigate the image captioning capabilities of several mainstream multimodal large language models (MLLMs). Meanwhile, we introduce image corruption benchmark tests to simulate common underwater degradation scenarios (turbidity, severe weather) and assess the robustness of vision models, enabling trustworthy decisions on ecological protection in the wild. ShellfishNet is dedicated to providing a data foundation and a model-evaluation benchmark for the intelligent monitoring of benthic organisms.
Authors: Haotian Zhang, Xu Mo, Yixin Yu, Guanhua Zhu, Jian Xue, Tongda Xu, Yan Wang, Jiaqi Zhang, Siwei Ma, Wen Gao
Abstract: Free-Viewpoint Video (FVV) has emerged as a cornerstone of next-generation immersive media systems and attracted widespread attention. Previous methods primarily focus on short video sequences and suffer from significant performance degradation when processing long-horizon free-viewpoint video (LFVV). Motivated by bit allocation theory, we analyze dynamic-anchor-based volumetric video representation within a rate-distortion optimization framework and propose \textbf{SoLAR}, which is the first error-resilient streamable FVV framework that maintains stable reconstruction quality on long sequences without requiring group-of-pictures partitioning. We propose the Anchor Activation Dynamics (AAD), which enables dynamic anchors to model non-rigid transformations by dynamically activating informative anchors and suppressing redundant ones. Furthermore, we introduce Latent Discrepancy Aware Recalibration (LaDAR), which is a mechanism to identify discrepancies between latent representations and recalibrate the correspondences encoded in the network, effectively mitigating error propagation in LFVV without compromising real-time performance or storage compactness. Extensive experiments demonstrate that \textbf{SoLAR} achieves state-of-the-art reconstruction performance while maintaining minimum storage overhead, which provides a new direction for LFVV reconstruction and advances the practical deployment of immersive systems. Demo free-viewpoint videos are provided in the supplementary material.
Authors: Miso Lee, Sangeek Hyun, Yerim Jeon, Jae-Pil Heo
Abstract: While Gaussian Splatting-based Feature Fields (GSFFs) have shown promise for visual localization, this paper highlights that photometrically optimized GSFFs are inherently ill-suited for 2D-3D matching. The volumetric extent of each Gaussian induces many-to-one pixel-to-point mappings that destabilize PnP-based pose estimation, while photometric optimization gives rise to superfluous Gaussians devoid of multi-view consistency. To address these issues, we propose SplitGS-Loc, a localization-specialized GSFFs construction framework that disambiguates 2D-3D correspondences by exploiting Gaussian attributes. Our key design, Mixture-of-Gaussians-based splitting, decomposes each Gaussian into smaller Gaussians, replacing ambiguous many-to-one with precise one-to-one correspondences. In parallel, we exploit composition weights from GS rasterization to select Gaussians that significantly and consistently contribute across multiple views and aggregate discriminative features through strong pixel-Gaussian associations, enforcing multi-view consistency. The resulting compact yet discriminative feature fields enable stable PnP convergence, achieving state-of-the-art performance on localization benchmarks. Extensive experiments validate that SplitGS-Loc extends the utility of photometric GSFFs to accurate and efficient localization by exploiting Gaussian attributes, without per-scene training or iterative pose refinement.
Authors: Simin Huo, Ning LI
Abstract: Video-language models (VLMs) face rapid inference costs as visual token counts scale with video length. For example, 32 frames at $448{\times}448$ resolution already yield >8,000 visual tokens in Qwen3-VL, making LLM prefill the dominant throughput bottleneck. Existing methods often rely on global similarity or attention-guided compression, incurring offsets to their gains. We propose \textbf{Temporal Token Fusion (TTF)}, a training-free, plug-and-play pre-LLM token compression framework that exploits structured temporal redundancy in video. TTF automatically selects an anchor frame, then for each subsequent frame, performs a local window similarity search (e.g.,$3\times 3$), fusing tokens that exceed a threshold. The compressed sequence maintains positional consistency across both prefill and decoding through coordinate realignment, enabling seamless integration with existing VLM pipelines. On Qwen3-VL-8B with threshold t=0.70, TTF removes about 67\% of visual tokens while retaining 99.5\% of the baseline accuracy and introducing only ${\approx}0.16$\,GFLOPs of matching overhead. Overall, TTF offers a practical, efficient solution for video understanding. The code is available at \href{https://github.com/Cominder/ttf}{https://github.com/Cominder/ttf}
URLs: https://github.com/Cominder/ttf, https://github.com/Cominder/ttf
Authors: Shuai Zhang, Zhecheng Shi, Zhuxiao Li, Jing Ou, Tengxi Wang, Yuan Liu, Wufan Zhao
Abstract: Semantic segmentation of large-scale 3D point clouds is crucial for applications such as autonomous driving and urban digital twins. However, the sparse sampling pattern of LiDAR and the view-dependent geometric distortion in image observations complicate cross-modal alignment and hinder stable fusion. Inspired by the fact that 2D images captured by cameras are representations of the 3D world, we recognize that the features learned from 2D and 3D segmentation share some common semantics, while other aspects remain modality-specific. This insight motivates a unified multimodal framework for joint 2D-3D semantic segmentation. We combine a SAM-based vision encoder with a SPTNet-based geometric encoder to extract complementary semantic and geometric representations. The resulting features from both modalities are explicitly decomposed into shared and private subspaces, where the shared components summarize semantic factors common to both domains, and the private components preserve properties that are unique to each modality. A lightweight attention-based fusion module aggregates the shared features into a consistent cross-modal representation, and a regularized training objective ensures both semantic alignment and subspace independence. Experiments on the SemanticKITTI and nuScenes benchmarks demonstrate consistent improvements in segmentation accuracy over representative multimodal baselines, accompanied by competitive computational efficiency. Cross-domain evaluation on nuScenes USA-Singapore shows stable performance under distribution shifts, demonstrating strong generalization. The implementation code is publicly available at: https://github.com/shuaizhang69/UniD-Shift.
Authors: Hanxi Li, Yao Cheng, Bo Zhang, Li Zeng
Abstract: Compared to RGB images, raw sensor data provides a richer representation of information, which is crucial for accurate recognition, particularly under challenging conditions such as low-light environments. The traditional Image Signal Processing (ISP) pipeline generates visually pleasing RGB images for human perception through a series of steps, but some of these operations may adversely impact the information integrity by introducing compression and loss. Furthermore, in computer vision tasks that directly utilize raw camera data, most existing methods integrate minimal ISP processing with downstream networks, yet the resulting images are often difficult to visualize or do not align with human aesthetic preferences. This paper proposes UniISP, a novel ISP framework designed to simultaneously meet the requirements of both human visual perception and computer vision applications. By incorporating a carefully designed Hybrid Attention Module (HAM) and employing supervised learning, the proposed method ensures that the generated images are visually appealing. Additionally, a Feature Adapter module is introduced to effectively propagate informative features from the ISP stage to subsequent downstream networks. Extensive experiments demonstrate that our approach achieves state-of-the-art performance across various scenarios and multiple datasets, proving its generalizability and effectiveness.
Authors: Xin Chen, Chuanyu Sun, Jiao Xu, Houwen Peng, Dong Wang, Huchuan Lu, Kede Ma
Abstract: Conventional visual object trackers localize targets using handcrafted spatial priors, often in the form of heatmaps. Such priors provide only surrogate supervision and are poorly aligned with tracking optimization and evaluation metrics, such as intersection over union (IoU) and area under the success curve (AUC). Here, we introduce RELO, a REinforcement-learning-to-LOcalize method for visual object tracking that formulates target localization as a Markov decision process. Specifically, RELO replaces handcrafted spatial priors with a localization policy learned over spatial positions via reinforcement learning, with rewards combining frame-level IoU and sequence-level AUC. We additionally introduce layer-aligned temporal token propagation to improve semantic consistency across frames, with negligible computational overhead. Across multiple benchmarks, RELO achieves superior results, attaining 57.5% AUC on LaSOText without template updates. This confirms that reward-driven localization provides an effective alternative to prior-driven localization for visual object tracking.
Authors: Yuyang Li, Jiashu Han, Yinyi Lai, Wenbin Kang, Zenghui Liu
Abstract: Marine debris detection for ocean robot is crucial for ecological protection, yet performance is often degraded by low-quality images with blur, complex backgrounds, and small targets. To address these challenges, we propose YOLO-MD, an enhanced YOLO-based detection framework. A Dual-Branch Convolutional Enhanced Self-Attention (DB-CASA) module is designed to strengthen spatial-channel interactions, improving feature representation in degraded images. Additionally, a lightweight shift-based operation is introduced to enhance fine-grained feature extraction for objects of varying scales while maintaining parameter efficiency. We further propose SFG-Loss to mitigate class imbalance and optimization instability via dynamic sample reweighting. Experiments on the UODM dataset demonstrate that YOLO-MD achieves 0.875 precision, 0.822 F1-score, and 0.849 mAP50, outperforming the latest state-of-the-art methods. The effectiveness of this method has also been verified through real-world robotic edge deployment experiments.
Authors: Haonan Wang, Hanyu Zhou, Tao Gu, Luxin Yan
Abstract: Generative models have achieved success in producing apparently coherent 2D videos, but remain challenging in the physical world due to lack of 4D spatiotemporal scale. Typically, existing 4D generative models directly embed macro scale constraints to enhance overall spatiotemporal consistency. However, these methods only ensure global appearance coherence and fail to reveal the local dynamics of the physical world. Our insight is that global appearance structure and local dynamic topology empower 4D spatiotemporal cognition, thereby enabling 4D generation with spatiotemporal regularities. In this work, we propose ST-Gen4D, a 4D generation framework with 4D spatiotemporal cognition-based world model. Our model is guided by four key designs: 1) Spatiotemporal representation. We encode various modalities into multiple representations as a feature basis. 2) Spatiotemporal cognition. We sculpture these representations into global appearance graph and local dynamic graph, and fuse them via semantic-bridged spatiotemporal fusion to obtain a 4D cognition graph. 3) Spatiotemporal reasoning. We utilize a world model to derive future state based on the 4D cognition. 4) Spatiotemporal generation. We leverage the derived cognition as condition to guide latent diffusion for 4D Gaussian generation. By deeply integrating 4D intrinsic cognition with generative priors, our model guarantees the structural rationality and topological consistency of 4D generation. Moreover, we propose ST-4D datasets by aggregating public 4D datasets and self-built subset. Extensive experiments demonstrate the superiority of our ST-Gen4D across 3D and 4D generation tasks.
Authors: Shaokai Ye, Vasileios Saveris, Yihao Qian, Jiaming Hu, Elmira Amirloo, Peter Grasch
Abstract: Image captioning is one of the most fundamental tasks in computer vision. Owing to its open-ended nature, it has received significant attention in the era of multimodal large language models (MLLMs). In pursuit of ever more detailed and accurate captions, recent work has increasingly turned to reinforcement learning (RL). However, existing captioning-RL methods and evaluation metrics often emphasize a narrow notion of caption quality, inducing trade-offs across core dimensions of captioning. For example, utility-oriented objectives can encourage noisy, hallucinated, or overlong captions that improve downstream question answering while harming fluency, whereas arena-style objectives can favor fluent but generic descriptions with limited usefulness. To address this, we propose a more balanced RL framework that jointly optimizes utility-aware correctness, reference coverage, and linguistic quality. In order to effectively optimize the resulting continuous multi-objective reward formulation, we apply GDPO-style reward-decoupled normalization to continuous-valued captioning rewards and show that it improves performance over vanilla GRPO. Additionally, we introduce length-conditional reward masking, yielding a more suitable length penalty for captioning. Across LLaVA-1.5-7B and Qwen2.5-VL 3B and 7B base models, our method consistently improves caption quality, with peak gains of +13.6 DCScore, +9.0 CaptionQA, and +29.0 CapArena across different models.
Authors: Zheyuan Gu, Minghao Shao, Zhen Wang, Yusong Wang, Mingkun Xu, Shijie Zhang, Hao Jiang
Abstract: While spatiotemporal deepfake detectors achieve high AUC, our experiments reveal their susceptibility to evasion attacks. These models tend to overfit on fragile temporal spectrum cues, rather than learning robust semantic causality. To mitigate this vulnerability, we propose SpInShield, a temporal spectral-invariant defense framework explicitly designed to decouple semantic motion from manipulatable spectral artifacts. We propose a learnable spectral adversary that dynamically synthesizes severe spectral deformations, simulating extreme attack scenarios. By employing a shortcut suppression optimization strategy, SpInShield compels the encoder to extract reliable forensic cues while purging unstable spectral statistics from the latent space. Experiments show that SpInShield obtains competitive performance on widely used datasets and outperforms the strongest baseline by 21.30 percentage points in AUC under simulated amplitude spectral attacks.
Authors: Yu Pan, Andi Zhang, Yi Wang, Sibei Yang, Wenjie Wang
Abstract: Diffusion Vision-Language Models (dVLMs), built upon the non-causal foundations of Diffusion Large Language Models (dLLMs), have demonstrated remarkable efficacy in multimodal tasks by departing from the traditional autoregressive generation paradigm. While dVLMs appear inherently robust against conventional jailbreak tactics, which we categorize as Fixed Prefix Optimization (FPO) (e.g., anchoring responses with "Sure, here is"), this perceived resilience is deceptive. Our investigation into the safety landscape of dVLMs reveals a unique refusal pattern: Immediate Refusal and Progressive Refusal. We find that while FPO-based attacks often fail by triggering the latter, the progressive refinement process itself uncovers a novel, latent attack surface. To exploit this vulnerability, we propose Global Probability Optimization (GPO), a general jailbreak paradigm designed specifically for the denoising trajectory of masked diffusion models. Unlike prefix-based methods, GPO manipulates the global generative dynamics to bypass guardrails in diffusion language models. Building on this, we introduce GPO-V, the first visual-modality jailbreak framework tailored for dVLMs. Empirical results demonstrate that GPO-V produces stealthy perturbations with exceptional cross-model transferability, revealing a critical security gap in non-sequential generative architectures. Our findings underscore the critical urgency of addressing safety alignment in dVLMs. These results necessitate an immediate and fundamental re-evaluation of current defense paradigms to mitigate the unique risks of diffusion-based generation. Our code is available at: https://anonymous.4open.science/r/GPO-V-0250.
Authors: Jie Li, Shulian Zhang, Yangyang Gao, Wenbo Li, Yulun Zhang, Yong Guo, Jian Chen
Abstract: Human insertion aims to naturally place specific individuals into a target background. Although existing image editing models may have such ability, they often produce failure cases, including inappropriate human pose in new background, inconsistent number of people, and modified facial identity. Moreover, publicly available human datasets often lack full-body portraits and realistic physical interaction between humans and their background. To address these challenges, we propose InsHuman for natural and identity-preserving human insertion. Specifically, we propose Human-Background Adaptive Fusion (HBAF), which detects foreground humans to obtain a binary mask and applies region-aware weighting to align the human regions between predicted and ground-truth latents, ensuring the person's pose, count, and overall appearance are coherently adapted to the target background.We further propose Face-to-Face ID-Preserving (FFIP), which detects and matches faces between the generated image and the source image in terms of face recognition features to enforce identity consistency for each face.In addition, we propose Bidirectional Data Pairing (BDP) strategy to construct BDP-InsHuman, a high-quality dataset with realistic human-background interactions. Experiments demonstrate that InsHuman achieves significant improvements in generating plausible images while keeping human identity unchanged.
Authors: Tianhao Niu, Ziyu Han, Qingfu Zhu, Wanxiang Che
Abstract: Referring expression grounding is a core problem in visual grounding and is widely used as a diagnostic of spatial grounding and reasoning in vision and language models, yet most prior work focuses on natural images. In contrast, existing chart referring expression grounding-related benchmarks remain limited: (1) they largely adopt bounding boxes, constraining localization precision for fine chart elements (2) they mostly assume a single and two referred target instances, failing to handle multi-instance target references; (3) the language expressions over-rely on textual cues or data-rank clues (4) they cover only a narrow range of chart types. To address these issues, we introduce a chart referring expression grounding benchmark that systematically supports multiple localization forms, multiple referred targets, diverse grounding cues and diverse chart types. Results across representative multimodal large models reveal a significant performance gap. We further introduce a code-driven synthesis pipeline that exploits the inherent alignment between plotting programs and rendered chart primitives to derive pixel accurate instance masks across chart element types and granularities. We train an instance segmentation model with the synthesized masks and integrate it into a general-purpose multimodal grounding framework. The resulting system consistently outperforms baselines on our benchmark and generalizes well to a ChartQA-derived real-chart grounding benchmark.
Authors: Yuanyan Li, Matthias Althoff
Abstract: Monocular depth estimation (MDE) typically produces depth estimations that are defined up to an unknown scale or shift. When only sparse metric anchors are available, recovering accurate metric depth becomes challenging yet necessary for practical applications. We address this problem by formulating metric depth recovery as image-adaptive scale field modeling. Instead of directly correcting the depth, we reformulate the correction as a low-dimensional linear combination of image-adaptive basis maps. These maps are derived from semantic and geometric cues encoded in the MDE estimations and intermediate representations. The weights of basis maps are efficiently determined from sparse metric anchors via a least-squares problem. This formulation yields improved metric depth accuracy, strong robustness under extreme anchor sparsity, and an interpretable decomposition of spatial scale variations. Extensive experiments across multiple datasets and representative MDE models demonstrate the effectiveness and general applicability of our approach.
Authors: Linxiao Shi, Siming Zheng, Zerong Wang, Hao Zhang, Jinwei Chen, Bo Li, Shifeng Chen, Peng-Tao Jiang
Abstract: Existing mobile devices are constrained by compact optical designs, such as small apertures, which make it difficult to produce natural, optically realistic bokeh effects. Although recent learning-based methods have shown promising results, they still struggle with photos captured under high digital zoom levels, which often suffer from reduced resolution and loss of fine details. A naive solution is to enhance image quality before applying bokeh rendering, yet this two-stage pipeline reduces efficiency and introduces unnecessary error accumulation. To overcome these limitations, we propose MagicBokeh, a unified diffusion-based framework designed for high-quality and efficient bokeh rendering. Through an alternative training strategy and a focus-aware masked attention mechanism, our method jointly optimizes bokeh rendering and super-resolution, substantially improving both controllability and visual fidelity. Furthermore, we introduce degradation-aware depth module to enable more accurate depth estimation from low-quality inputs. Experimental results demonstrate that MagicBokeh efficiently produces photorealistic bokeh effects, particularly on real-world low-resolution images, paving the way for future advancements in bokeh rendering. Our code and models are available at https://github.com/vivoCameraResearch/MagicBokeh.
Authors: Hao Wang, Yiqun Sun, Pengfei Wei, Lawrence B. Hsieh, Daisuke Kawahara
Abstract: Vision-language models (VLMs) have advanced rapidly and are increasingly deployed in real-world applications, especially with the rise of agent-based systems. However, their safety has received relatively limited attention. Even the latest proprietary and open-weight VLMs remain highly vulnerable to adversarial attacks, leaving downstream applications exposed to significant risks. In this work, we propose a novel and lightweight adversarial attack detection framework based on sparse autoencoders (SAEs), termed SAEgis. By inserting an SAE module into a pretrained VLM and training it with standard reconstruction objectives, we find that the learned sparse latent features naturally capture attack-relevant signals. These features enable reliable classification of whether an input image has been adversarially perturbed, even for previously unseen samples. Extensive experiments show that SAEgis achieves strong performance across in-domain, cross-domain, and cross-attack settings, with particularly large improvements in cross-domain generalization compared to existing baselines. In addition, combining signals from multiple layers further improves robustness and stability. To the best of our knowledge, this is the first work to explore SAE as a plug-and-play mechanism for adversarial attack detection in VLMs. Our method requires no additional adversarial training, introduces minimal overhead, and provides a practical approach for improving the safety of real-world VLM systems.
Authors: Lan Chen, Qi Mao, Yiren Song, Yuchao Gu, Siwei Ma
Abstract: Visual-prompt-guided edit transfer aims to learn image transformations directly from example pairs, offering more precise and controllable editing than purely text-driven approaches. However, existing diffusion transformer-based methods often fail to faithfully reproduce the demonstrated edits due to structural mismatches between the task and the backbone, including a pretrained bias toward textual conditioning and inherent stochastic instability during sampling. To bridge this gap, we present EditTransfer++, a framework that combines progressively structured training with an efficient conditioning scheme to improve both visual prompt faithfulness and inference efficiency. We first mitigate textual dominance with a text-decoupled training strategy that removes text conditioning during fine-tuning, compelling the model to infer transformations solely from visual evidence while still supporting optional text guidance at inference. On top of this visually grounded model, a best-worst contrastive refinement mechanism reshapes the denoising trajectories to suppress unfaithful generations and improve consistency across random seeds. To alleviate the computational bottleneck of high-resolution in-context editing, we further introduce a condition compression and reuse strategy that reduces token redundancy and enables efficient generation of images with a 1024-pixel long edge. Extensive experiments on existing benchmarks and the proposed EditTransfer-Bench show that EditTransfer++ achieves state-of-the-art visual prompt faithfulness with substantially faster inference than prior methods, suggesting a promising direction for scalable prompt-guided image editing and broader visual in-context learning.
Authors: Zitong Xu, Huiyu Duan, Yifei Nie, Mingda Du, Sijing Wu, Xiongkuo Min, Tianyi Zheng, Jian Zhang, Shusong Xu, Jinwei Chen, Bo Li, Guangtao Zhai
Abstract: Recent text-guided image editing (TIE) models have made remarkable progress, yet edited images still frequently suffer from fine-grained issues such as unnatural objects, lighting mismatch, and unexpected changes. Existing refinement approaches either rely on costly iterative regeneration or employ vision-language models (VLMs) with weak spatial grounding, often resulting in semantic drift and unreliable local corrections. To address these limitations, we first construct EditFHF-15K, a dataset of fine-grained human feedback for edited images, comprising (1) 15K images from 12 TIE models spanning 43 editing tasks, (2) 60K annotated artifact regions and 80K editing failure regions, each accompanied by textual reasoning, and (3) 45K mean opinion scores (MOSs) assessing perceptual quality, instruction following, and visual consistency. Based on EditFHF-15K, we propose EditRefiner, a hierarchical, interpretable, and human-aligned agentic framework that reformulates post-editing correction as a human-like perception-reasoning-action-evaluation loop. Specifically, we introduce: (1) a perception agent that detects contextual saliency maps of artifacts and editing failures, (2) a reasoning agent that interprets these perceptual cues to perform human-aligned diagnostic inference, (3) an action agent that uses the reasoning output to plan and execute localized re-editing, and (4) an evaluation agent that assesses the re-edited image and guides the action agent on whether further refinements are required. Extensive experiments demonstrate that EditRefiner consistently outperforms state-of-the-art methods in distortion localization, diagnose accuracy and human perception alignment, establishing a new paradigm for self-corrective and perceptually reliable image editing. The code is available at https://github.com/IntMeGroup/EditRefiner.
Authors: Ioan-Tudor-Alexandru Anghel, Ciprian-Mihai Ceausescu, Elena Dana Nedelcu, Elena Raluca Stirban, Camelia Croitoru, Despina Ungureanu, Ana Maria Palan, Gabriela Pop
Abstract: Non-alcoholic fatty pancreas disease (NAFPD) is an underdiagnosed condition associated with metabolic syndrome, insulin resistance, and increased risk of pancreatic cancer. Diagnosis typically relies on subjective visual assessment of ultrasound images by clinicians. We propose an end-to-end framework for automatically classifying normal versus fatty pancreas from abdominal ultrasound images. Our method employs a TransUNet-based segmentation architecture with a ResNet encoder and transformer bottleneck to delineate the pancreas and the splenic vein, followed by anatomically-guided patch extraction and patient-level classification through pairwise texture comparison. The feature engineering mimics clinical reasoning by comparing the echogenicity of peri-venous fat to the pancreatic parenchyma, providing an interpretable signal for classification. The segmentation models are initialized via domain-specific transfer learning from a liver segmentation task. We validate the full pipeline on a clinical dataset of 214 abdominal ultrasound images with 107 expert-labeled cases using 5-fold cross-validation. SVM with RBF kernel achieves a mean cross-validated accuracy of 89.7\%\,$\pm$\,1.8\% and F1 of 0.898\,$\pm$\,0.019, while the unsupervised K-Means baseline reaches 87.8\% accuracy, demonstrating that the proposed features capture the relevant clinical signal even without labeled training data. To our knowledge, this is the first end-to-end automated framework for fatty pancreas classification from ultrasound using segmentation-guided texture analysis.
Authors: Yuhao Zhou, Yunpeng Zhu, Yang Zhou, Jindi Lyu, Jian Lan, Zhangyuan Wang, Dan Si, Thomas Seidl, Qing Ye, Jiancheng Lyu
Abstract: Vision-Language-Action (VLA) models hold great promise for general-purpose robotic intelligence, yet scaling up such models is severely bottlenecked by the high cost of acquiring annotated training data. Fortunately, vision-equipped robots deployed across various domains already produce abundant vision-action pairs that can be leveraged to scale up VLA training more efficiently. However, these raw data cannot be centrally aggregated due to various constraints and also exhibit severe heterogeneity. To address these challenges, in this paper, we propose ForgeVLA, a federated VLA training framework that learns VLA models from distributed vision-action pairs without centralizing raw data or requiring manual annotations. Specifically, each client in ForgeVLA is equipped with an embodied instruction classifier that maps vision-action pairs to a predefined instruction set, recovering the missing language modality and forming complete vision-language-action triplets. Beyond triplet construction, we also identify vision-language feature collapse as a critical challenge that has been largely overlooked in prior federated VLA research. To mitigate this issue, ForgeVLA combines a client-side contrastive planning loss with a server-side adaptive aggregation strategy to learn task-discriminative representations efficiently. Extensive experiments across multiple benchmarks show that ForgeVLA significantly outperforms other baselines, and ablation studies further validate the contribution of each component.
Authors: Honghua Chen, Zitong Xu, Huiyu Duan, Xinyun Zhang, Xiongkuo Min, Guangtao Zhai
Abstract: Recent text-guided image editing (TIE) models have achieved remarkable progress, however, many edited results still suffer from artifacts, unintended modifications, and suboptimal aesthetics. Although several benchmarks and evaluation methods have been proposed, most existing approaches rely on scalar scores and lack interpretability. This limitation largely stems from the absence of high-quality interpretation datasets for TIE and effective reward models to train interpretable evaluators. To address these challenges, we introduce ReasonEdit-22K, the first dataset that combines 22K edited images with 113K Chain-of-Thought (CoT) samples, along with 1.3M human judgments assessing these interpretations in terms of logicality, accuracy, and usefulness. Building upon this dataset, we propose RE-Reward, a multimodal large language model (MLLM)-based reward model designed to provide human-aligned feedback for evaluating interpretable reasoning in image editing. Furthermore, we develop ReasonEdit, which is trained using reward signals derived from RE-Reward and the Group Relative Policy Optimization (GRPO) algorithm to learn an interpretable evaluation model. Extensive experiments demonstrate that ReasonEdit achieves superior alignment with human preferences and exhibits strong generalization across public benchmarks. In addition, it is capable of generating high-quality interpretable evaluation text, enabling more transparent and trustworthy assessment for image editing. The code is available at https://github.com/IntMeGroup/ReasonEdit.
Authors: Kai Zheng, Zejian Kang, Rui Mao, Hongyuan Zou, Yuanchen Fei, Xuanyang Xu, Xiangru Huang
Abstract: Speech-driven facial animation requires accurate correspondence between acoustic signals and facial motion, especially for articulation-related mouth movements. However, directly mapping speech audio to facial coefficients often overlooks the linguistic and phonetic structure underlying speech production. In this paper, we propose AudioFace, a language-assisted framework for speech-driven blendshape generation that treats mouth-related facial coefficient prediction as a structured generation problem guided by linguistic and articulatory information. Instead of relying solely on acoustic features, our method leverages the prior knowledge of multimodal large language models and introduces transcript- and phoneme-level cues to bridge speech signals with interpretable facial actions. Extensive experiments show that AudioFace achieves superior performance across multiple evaluation metrics, validating the effectiveness of language-assisted and multimodal-prior-guided speech-driven facial animation.
Authors: Ivan De Boi, Bart Ribbens, Veronika Golanova, Ursula Kapov, Simon Verspeek
Abstract: This paper proposes a novel framework for implicit multi-camera system calibration utilizing Gaussian Process (GP) regression. Conventional explicit calibration methods are constrained by rigid mathematical models and struggle with complex, non-linear distortions from unconventional optics, while existing neural network-based implicit approaches are typically data-hungry and lack inherent uncertainty quantification (UQ). Our GP-based model directly learns the complex, non-linear mapping from 2D image coordinates across all cameras to a 3D world coordinate, completely bypassing time-consuming estimation of explicit intrinsic and extrinsic parameters. Moreover, the inherent UQ is critical for transforming a simple 3D point prediction into a verifiable 3D measurement, complete with statistically-sound confidence bounds. To further enhance data efficiency and practical deployment, we integrate Active Learning (AL), which intelligently leverages the GP's predictive uncertainty to strategically guide the acquisition of new calibration data. This approach results in a robust, data-efficient, and reliable calibration solution, proving particularly effective in practical scenarios where collecting extensive calibration data is a dominant constraint. Our experiments show that the uncertainty for the 3D predictions is higher closer to the cameras. The data points in $uv$-coordinate space are more sparse in that region, even though they are not in 3D space. This work is relevant for anyone who is tasked with the calibration of complex multi-camera systems.
Authors: Zhiheng Li, Zongyang Ma, Jiaxian Chen, Jianing Zhang, Zhaolong Su, Yutong Zhang, Zhiyin Yu, Ruiqi Liu, Xiaolei Lv, Bo Li, Jun Gao, Ziqi Zhang, Chunfeng Yuan, Bing Li, Weiming Hu
Abstract: The past year has seen over 20 open-source document parsing models, yet thefield still benchmarks almost exclusively on OmniDocBench, a 1,355-pagemanually annotated dataset whose top scores have saturated above 90%. Athree-stage audit pipeline we run on OmniDocBench screens its 21,353evaluator-scored blocks and confirms 2,580 errors (12.08%); combined with overa year of public availability, both annotation quality and contamination riskcall its rankings into question. To address these issues, we presentPureDocBench, a programmatically generated, source-traceable benchmark thatrenders document images from HTML/CSS and produces verifiable annotations fromthe same source, covering 10 domains, 66 subcategories, and 1,475 pages, eachin three versions: clean, digitally degraded, and real-degraded (4,425 imagestotal). Evaluating 40 models spanning pipeline specialists, end-to-endspecialists, and general-purpose VLMs, we find: (i) document parsing is farfrom solved: the best model scores only ~74 out of 100, with a 44.6-point gapbetween the strongest and weakest models; (ii) specialist parsers with <=4Bparameters rival or surpass general VLMs that are 5-100x larger, yet formularecognition remains a shared bottleneck where no model exceeds 67% whenaveraging the formula metric across all three tracks; (iii) general VLMs loseonly 0.99/8.52 Overall points under digital/real degradation versus 4.90/14.21for pipeline specialists, producing ranking reversals that make clean-onlyevaluation misleading for deployment. All data, code, and artifacts arepublicly released.
Authors: Mengxin Qin, Xiang Zhang, Xi Wang, Kun Wei, Xu Yang, Cheng Deng
Abstract: Continual learning enables vision-language models to accumulate knowledge and adapt to evolving tasks without retraining from scratch. However, in multi-domain task-incremental learning, large domain shifts intensify the stability-plasticity dilemma. Most existing methods rely on fixed architectures with statically allocated parameters, which limits adaptation to new domains and aggravates catastrophic forgetting. To address these challenges, we propose DIMoE-Adapters, a Dynamic Incremental Mixture-of-Experts Adapters framework that introduces a dynamic expert evolution paradigm to balance stability and plasticity. This paradigm is implemented through two collaborative components: Self-Calibrated Expert Evolution (SCEE) and Prototype-Guided Expert Selection (PGES). SCEE constructs and evolves a sparse expert pool through expert optimization dynamics, improving plasticity while reducing redundant capacity. PGES controls expert utilization based on the pool shaped by SCEE, improving stability across both previously encountered and unseen tasks. Extensive experiments show that DIMoE-Adapters outperforms previous state-of-the-art methods across various settings.
Authors: Yujin Cho, Flavien Armangeon, Yanhao Li
Abstract: Unpaired smartphone ISP is a challenging problem due to the lack of scene and color alignment between RAW and target RGB images. Many existing methods either require paired data or rely heavily on adversarial training, which can become unstable in the unpaired setting. In this work, we present a simple and effective approach developed for the NTIRE 2026 Learned Smartphone ISP Challenge with Unpaired Data. Our method first reconstructs larger images from training patches to recover global context. Then, we extract semantic embeddings with DINOv2, and use fused Gromov-Wasserstein (FGW) optimal transport to build pseudo pairs between RAW and RGB images at both image and patch levels. This semantic matching allows us to partially alleviate the unpairedness of the data and build these pseudo input-target pairs. Based on these pseudo pairs, we train a lightweight CNN with only 7K parameters for color rendering. The network is designed to be compact and focus on color transformation rather than structural change, which helps reduce artifacts and improve training stability. Our challenge submission achieves 22.569 PSNR, 0.675 SSIM, and 8.067 $\Delta E$ on the final hidden test set, significantly improving over the baseline and achieving the 3rd best SSIM and $\Delta E$ among all challenge entries. Our code is available at github.com/nuniniyujin/Unpaired-ISP .
Authors: Tianchi Xu, Ziqiang Ma, Andrea Marinoni, Yuanpeng He, Xiaoqing Li, Chuanfeng Zhao, Kang He, Jintao Xu, Bohan Zhou, Wenbo Zhao, Haoshuang Chen, Tun Wang, Dongdong Wang, Yang Hong
Abstract: Accurate four-dimensional (4D) precipitation information is essential for understanding the Earth's energy and water cycles, yet remains observationally unresolved at global scales. Conventional theory holds that geostationary infrared observations primarily sense cloud-top properties, with limited sensitivity to sub-cloud precipitation. Here we show that cloud-top infrared measurements nevertheless encode sufficient information to recover the four-dimensional structure of precipitation, revealing a previously unexploited observability of sub-cloud processes. We introduce a physically constrained deep learning framework, 4DPrecipNet, in which a moisture-first constraint requires the latent representation to recover precipitable water vapour, anchoring the model in thermodynamic consistency. By integrating multi-channel infrared radiances with these constraints and radar-derived precipitation profiles, we reconstruct the vertical and temporal evolution of precipitation systems from geostationary orbit. The framework captures deep convective structures and their evolution, with robust performance across large samples and independent radar comparisons. These results demonstrate that sub-cloud precipitation is physically encoded in cloud-top infrared observations, establishing a new pathway for continuous global monitoring of precipitation structure.
Authors: Jingyuan Zhu, Biaolong Chen, Le Zhang, Aixi Zhang, Hao Jiang, Pipei Huang
Abstract: Efficiently aligning large-scale video diffusion models with human intent requires a scalable and trajectory-aware pathway that bridges the inherent discrepancy between training noise distributions and practical inference trajectories. While existing paradigms such as Direct Preference Optimization (DPO) and Group Relative Policy Optimization (GRPO) attempt to address this, they are often hindered by either reliance on bias-prone, complex reward models or suboptimal timestep sampling. In this paper, we propose Diffusion-APO (Aligned Preference Optimization), a trajectory-aware algorithm that resolves this misalignment by synchronizing training noise with inference-time denoising paths to maximize gradient signal efficacy. To translate this algorithmic innovation into a practical solution, we introduce a unified and modular RLHF framework that integrates online ranking, half-online anchoring, offline refinement, and distillation-aware drift correction. This framework enables flexible, multi-stage preference alignment across diverse data and computational constraints without relying on scalar-reward-based policy gradients. Through extensive experiments, we demonstrate that Diffusion-APO consistently outperforms standard baselines in visual quality and instruction following, while effectively preserving generative fidelity during model acceleration, providing a robust, end-to-end pathway for scalable video diffusion alignment.
Authors: Bohan Hou, Jiuning Gu, Jiayan Guo, Ronghao Dang, Sicong Leng, Xin Li, Xuemeng Song, Jianfei Yang
Abstract: Existing benchmarks for multimodal agentic search evaluate multimodal search and visual browsing, but visual evidence is either confined to the input or treated as an answer endpoint rather than part of an interleaved search trajectory. We introduce \textbf{InterLV-Search}, a benchmark for Interleaved Language-Vision Agentic Search, in which textual and visual evidence is repeatedly used to condition later search. It contains 2,061 examples across three levels: active visual evidence seeking, controlled offline interleaved multimodal search, and open-web interleaved multimodal search. Beyond existing benchmarks, it also includes multimodal multi-branch samples that involve comparison between multiple entities during the evidence search. We construct Level 1 and Level 2 with automated pipelines and Level 3 with a machine-led, human-supervised open-web pipeline. We further provide InterLV-Agent for standardized tool use, trajectory logging, and evaluation. Experiments on proprietary and open-source multimodal agents show that current systems remain far from solving interleaved multimodal search, with the best model below 50% overall accuracy, highlighting challenges in visual evidence seeking, search control, and multimodal evidence integration. We release the benchmark data and evaluation code at https://github.com/hbhalpha/InterLV-Search-Bench
Authors: Mengxin Qin, Xiang Zhang, Kun Wei, Xu Yang, Cheng Deng
Abstract: Class-incremental learning aims to continuously acquire new knowledge while preserving previously learned information, thereby mitigating catastrophic forgetting. Existing methods primarily restrict parameter updates but often overlook their structural properties in high-dimensional spaces. From a subspace perspective, updates induced by different tasks tend to lie in multiple overlapping low-rank subspaces, leading to cross-task subspace interference and severe forgetting. To address this issue, we propose HDSD, a Hierarchical Dual-Subspace Decoupling framework for continual learning in vision-language models. Specifically, we introduce a lightweight Feature Modulation Module (FMM) that explicitly decomposes the parameter space into general and task-specific subspaces. Building on this design, we develop two complementary components. First, a General Fusion Module (GFM) evaluates relative parameter changes across tasks and uses an adaptive threshold to capture stable and transferable knowledge. Second, a Hierarchical Learning Module (HLM) performs structured parameter decomposition via Singular Value Decomposition (SVD) and uses a scaling mechanism to constrain updates within distinct subspace scales. Together, these designs reduce subspace interference and parameter drift. Extensive experiments on conventional benchmarks show that HDSD achieves state-of-the-art results.
Authors: Yuanzhi Wang, Xuhua Ren, Jiaxiang Cheng, Bing Ma, Kai Yu, Tianxiang Zheng, Qinglin Lu, Zhen Cui
Abstract: Human image animation has witnessed significant advancements, yet generating high-fidelity hand motions remains a persistent challenge due to their high degrees of freedom and motion complexity. While reinforcement learning from human feedback, particularly direct preference optimization, offers a potential solution, it necessitates the construction of strict preference pairs. However, curating such pairs for dynamic hand regions is prohibitively expensive and often impractical due to frame-wise inconsistencies. In this paper, we propose Implicit Preference Alignment (IPA), a data-efficient post-training framework that eliminates the need for paired preference data. Theoretically grounded in implicit reward maximization, IPA aligns the model by maximizing the likelihood of self-generated high-quality samples while penalizing deviations from the pretrained prior. Furthermore, we introduce a Hand-Aware Local Optimization mechanism to explicitly steer the alignment process toward hand regions. Experiments demonstrate that our method achieves effective preference optimization to enhance hand generation quality, while significantly lowering the barrier for constructing preference data. Codes are released at https://github.com/mdswyz/IPA
Authors: Christopher Ries, Moussa Kassem Sbeyti, Nicolas Bianco, Nadja Klein
Abstract: Conformal Prediction (CP) is a distribution-free method for constructing prediction sets with marginal finite-sample coverage guarantees, making it a suitable framework for reliable uncertainty quantification in safety-critical object detection. However, object detection introduces structured multi-output predictions, complicating the application of classical CP theory developed for single outputs. In addition, standard, unscaled CP produces fixed-width prediction intervals across inputs, leading to unnecessary width for low-uncertainty predictions. While scaled CP addresses this by adapting the interval width to an input-dependent uncertainty estimate, prior work has neither systematically compared unscaled and scaled CP for multi-class object detection, nor integrated CP with a complementary uncertainty quantification method in this setting. We fill this gap by: (i) applying CP coordinate-wise to bounding box corners with a Bonferroni correction for box-level guarantees; (ii) scaling the resulting intervals using per-prediction aleatoric uncertainty estimates derived from a probabilistic object detector trained with loss attenuation, evaluated in uncalibrated and two calibrated variants; (iii) extending to a two-step pipeline that constructs prediction sets for the class using RAPS and conditions the conformalized bounding boxes on the predicted class set. Across three autonomous driving datasets (KITTI, BDD, CODA), including a cross-domain setting under distribution shift, scaled CP consistently improves interval sharpness over unscaled CP, achieving up to 19% higher IoU and 39% lower interval scores, without sacrificing coverage. Class-wise calibration further improves coverage for both variants with a negligible effect on sharpness. Together, these improvements yield more actionable uncertainty estimates for real-time, real-world object detection.
Authors: Grzegorz Wilczynski, Miko{\l}aj Zielinski, Bartosz \'Swirta, Dominik Belter, Przemys{\l}aw Spurek
Abstract: 3D vision systems are fundamentally constrained by their reliance on visual overlap: reconstruction methods require it for geometric alignment, while generative models use it to enforce multi-view consistency. This limitation is particularly acute in real-world scenarios such as distributed swarm robotics or crowd-sourced data collection, where capturing overlapping perspectives, both in terms of spatial and appearance overlap, is often impossible. We introduce Generative Reconstruction from Disjoint Views as a new paradigm, establish a comprehensive dataset, and propose specialized evaluation metrics for zero-overlap scenarios. Our benchmarking demonstrates that existing state-of-the-art methods fail catastrophically on this task, producing disconnected geometries or semantically incoherent reconstructions. To address these limitations, we propose GLADOS, a general, modular framework that operates through three stages: (1) Generative Bridging, where foundation models synthesize intermediate perspectives to connect disjoint inputs; (2) Robust Coarse 3D Reconstruction, that establish coarse geometric scaffold via global alignment which absorbs local contradictions from generative process; and (3) Iterative Context Expansion and Consistency Optimization to fill missing regions and unify the reconstruction. As an architectureagnostic framework, GLADOS enables seamless integration of future advances in generation, reconstruction, and inpainting. The source code is available at: https://github.com/gwilczynski95/GLADOS.
Authors: Zepeng Yang, Junxuan Bai, Hao Li, Ju Dai, Junjun Pan, Yongfeng Yin, Bin Li
Abstract: The rapid advances in deep learning have significantly enhanced the accuracy of multimodal 3D human pose estimation (HPE). However, the state-of-the-art (SOTA) HPE pipelines still rely on Transformers, whose quadratic complexity makes real-time processing for long sequences impractical. Mamba addresses this issue through selective state-space modeling, enabling efficient sequence processing without sacrificing representational power. Nevertheless, it struggles to capture complex spatial dependencies in multimodal settings. To bridge this gap, we propose VIMCAN, a hybrid architecture that combines the efficient sequence modeling of Mamba with the spatial reasoning of Cross-Attention, and performs robust visual-inertial fusion and human pose estimation between RGB keypoints and wearable IMU data. By leveraging Mamba's dynamic parameterization for temporal modeling and Attention for spatial dependency extraction, VIMCAN achieves superior accuracy, with mean per-joint position errors (MPJPE) of 17.2 mm on TotalCapture and 45.3 mm on 3DPW. VIMCAN outperforms prior Transformer-based and other SOTA approaches while supporting real-time inference at over 60 frames per second on consumer-grade hardware. The source code is available on GitHub.
Authors: Nishant Suresh Aswani, Saif Eddin Jabari
Abstract: Recent work has shown that contiguous vision transformer (ViT) blocks (a) can be replaced by a linear map and (b) organize into recurrent phases of computation. We ask whether these observations coincide: does ViT depth implement approximately \textit{autonomous linear} dynamics, admitting a single operator $K$ applied recurrently across a contiguous span? We test this using Dynamic Mode Decomposition (DMD), which fits $K$ from selected, consecutive hidden-state pairs and predicts $p$ steps ahead via $K^p$. On four pretrained DINO ViTs, we study the regularization, rank, and calibration budget required for stable fitting. For short spans ($p \leq 4$), $K^p$ tracks an unconstrained endpoint map to within $0.02$ cosine similarity on DINOv3-H/16+, while also recovering intermediate activations at each skipped block. At early cut starts, the fitted operators compress to rank $\ll d$ with minimal calibration data, and across tokens, \texttt{cls} is most amenable to linearization; both properties decay monotonically with depth. Yet this local fidelity does not transfer downstream. At the final hidden state, after propagating through the remaining blocks, an identity baseline becomes competitive.
Authors: Alice Natalina Caragliano, Valerio Guarrasi, Michela Gravina, Carlo Sansone, Paolo Soda
Abstract: Pathological complete response (pCR) is a key prognostic factor in breast cancer patients undergoing neoadjuvant therapy, strongly associated with long-term survival and treatment personalization. However, accurate pre-treatment pCR prediction remains challenging due to severe class imbalance and limited generalizability across diverse clinical settings. In this work, we propose a multimodal stepwise clinically-guided attention learning framework for pCR prediction from breast magnetic resonance imaging (MRI), designed to address these limitations through medically grounded spatial guidance and multimodal integration. The approach follows a stepwise training strategy inspired by physician reasoning: the model first learns global discriminative imaging patterns, then attention mechanisms are introduced to constrain the network toward tumor regions, and finally clinical variables are integrated to refine decision-making. This guidance strategy encourages prioritization of task-relevant features, improving identification of responders despite their limited representation in the dataset. Moreover, grounding attention in anatomically consistent tumor regions reduces reliance on dataset-specific patterns, thereby enhancing cross-institutional generalization. The framework is evaluated through external validation across heterogeneous MRI cohorts. Compared to non-guided single-stage baselines, the proposed approach improves sensitivity while maintaining competitive specificity, and produces anatomically coherent attention maps that support interpretation of the model's predictions. These findings highlight the potential of clinically-guided multimodal attention learning for robust and generalizable pCR prediction in breast cancer.
Authors: Song Zhang, Yanlong Chen, Yilin Li, Yining Chen, Zili Yi, Xiaowei Zhang, Yawei Li
Abstract: Remote sensing vision-language models (RS-VLMs) face a fundamental mismatch with natural-image counterparts: the same geographic object exhibits radically different visual evidence across ground sampling distances (GSDs) spanning multiple orders of magnitude. Yet existing RS-VLMs often discard GSD or inject it as a discrete text token, forcing a single static parameter set to absorb the entire scale spectrum. We introduce ScaleEarth, a parameter-efficient fine-tuning framework built on Qwen3-VL that treats GSD as a continuous conditioning variable governing the model's computation path. At its core, CS-HLoRA (Continuous Scale-Conditioned Hyper-LoRA) modulates the LoRA low-rank subspace through a GSD-driven gate, enabling the model to dynamically route computation by physical scale. To remove reliance on sensor metadata at deployment, we pair CS-HLoRA with SSE-U, a lightweight heteroscedastic sub-head that predicts GSD and its uncertainty from visual features. To provide matching supervision, we construct GeoScale-VQA, a 1.5M-sample scale-layered RS-VQA corpus whose question-answer generation is conditioned on the same physical scalar that drives CS-HLoRA, forming a closed method-data loop. Trained with QLoRA on an 8B backbone, ScaleEarth achieves state-of-the-art results on remote-sensing benchmarks covering diverse Earth-system tasks, including XLRS-Bench and OmniEarth-Bench.
Authors: Peitao Han, Fei Cheng, Lis K. Pereira, Qianying Liu, Shigeru Kitazawa
Abstract: The Arrow-of-Time (AoT) task, determining whether a video plays forward or backward by recognizing temporal irreversibility, is one humans solve with near-perfect accuracy, yet frontier Video Large Language Models (Video-LLMs) perform only modestly above chance. This gap raises a key question: do visual backbones fail to encode temporal information, or does information bottleneck lie elsewhere in the Video-LLM architecture? We address this question by isolating the vision encoder from the Video-LLM and tracing temporal information across the encoder, projector, and LLM. We find that video-centric encoders with explicit temporal modeling encode strong temporal signals, whereas frame-centric encoders do not. However, when video-centric representations are passed through a standard Video-LLM architecture, performance often collapses, revealing a bottleneck of temporal information flow. We identify projector design as a key factor: Q-Former disrupts temporal information, while a time-preserved MLP projection substantially improves the LLM's access to such information. Our layer-wise analysis further shows temporal representation dynamics across encoder layers. Guided by these findings, we build a Video-LLM with temporal-aware video-centric encoder, time-preserved projector, and AoT supervision, surpassing human performance on AoT$_{PPB}$ with 98.1\% accuracy, and improving broader temporal reasoning tasks by up to 6.0 points on VITATECS-Direction and 1.3 points on TVBench. Our results show that temporal reasoning in Video-LLMs requires both effective temporal encoding and reliable transfer of this information to the LLM.
Authors: Yuliang Li, Chu Zhou, Heng Guo, Boxin Shi, Imari Sato, Zhanyu Ma
Abstract: Mainstream vision-language models (VLMs) fundamentally struggle with severe optical ambiguities, such as reflections and transparent objects, due to the inherent limitations of standard RGB inputs. While polarization imaging captures polarimetric physical parameters that resolve these ambiguities, existing methods are constrained by fixed-format outputs and remain isolated from open-ended reasoning. To bridge this semantic-physical gap, we introduce PolarVLM, the first multimodal framework integrating polarimetric physical parameters into VLMs. By employing a dual-stream architecture and a progressive two-stage training strategy, PolarVLM effectively prevents physical misinterpretations while preserving general visual abilities. Complementing our architecture, we construct PolarVQA, the first benchmark for polarization-aware VQA, featuring 75K physics-grounded instruction-tuning pairs targeting reflective and transparent scenes. Experiments show that PolarVLM surpasses the RGB baseline by 25.4% overall across five evaluation tasks, with remarkable gains of 26.6% in reflection recognition and 34.0% in glass counting, successfully unlocking physics-aware semantic understanding.
Authors: Ke Ma, Jiaqi Tang, Bin Guo, Xueting Han, Ruonan Xu, Qingfeng He, Ziheng Wang, Xu Wang, Qifeng Chen, Zhiwen Yu, Yunhao Liu
Abstract: Proactive streaming video understanding requires Video-LLMs to decide when to respond as a video unfolds, a task where existing methods often fall short due to their implicit, query-agnostic modeling of visual evidence. We introduce Response-G1, a novel framework that establishes explicit, structured alignment between the accumulated video evidence and the query's expected response conditions via scene graphs. The framework operates in three fine-tuning-free stages: (1) online query-guided scene graph generation from streaming clips; (2) memory-based retrieval of the most semantically relevant historical scene graphs; and (3) retrieval-augmented trigger prompting for per-frame "silence/response" decisions.By grounding both evidence and conditions in a shared graph representation, Response-G1 achieves more interpretable and accurate response timing decisions. Experimental results on established benchmarks demonstrate the superiority of our method in both proactive and reactive tasks, validating the advantage of explicit scene graph modeling and retrieval in streaming video understanding.
Authors: Pedro Alonso, Chongshou Li, Tianrui Li
Abstract: Despite extensive progress in point cloud robustness, existing methods primarily improve performance through augmentation or defense mechanisms, while overlooking the geometric root cause of adversarial fragility. We hypothesize that adversarial vulnerability in 3D networks arises from a manifold misalignment between the latent geometry learned by the model and the intrinsic geometry of the underlying surface. Small, geometry-preserving perturbations along the input manifold often induce disproportionate distortions in feature space, revealing a misalignment between latent and intrinsic geometries. We formalize this phenomenon by developing a geometric interpretation of 3D robustness that links classical adversarial theory to the intrinsic structure of point clouds. Motivated by this analysis, we introduce Manifold-Aligned Point Recognition (MAPR), a framework that regularizes the latent geometry by aligning predictions across intrinsic perturbations. MAPR augments each point cloud with intrinsic features capturing local curvature and diffusion structure, and applies a consistency loss that preserves invariance to intrinsic, geometry-preserving perturbations. Without relying on adversarial training or additional data, MAPR consistently improves robustness across multiple adversarial attacks on both the ModelNet40 and ScanObjectNN datasets, achieving average robustness gains of +20.02% and +8.58% on ModelNet40 and ScanObjectNN, respectively.
Authors: Hengyi Feng, Hao Liang, Mingrui Chen, Bohan Zeng, Meiyi Qiang, Zhengyang Zhao, Zimo Meng, Zeang Sheng, Wentao Zhang
Abstract: Real-world audio-visual understanding requires chaining evidence that is sparse, temporally dispersed, and split across the visual and auditory streams, whereas existing benchmarks largely fail to evaluate this capability. They restrict videos to short clips, isolate modalities, or reduce questions to one-hop perception. We introduce TraceAV-Bench, the first benchmark to jointly evaluate multi-hop reasoning over long audio-visual trajectories and multimodal hallucination robustness. TraceAV-Bench comprises 2,200 rigorously validated multiple-choice questions over 578 long videos, totaling 339.5 hours, spanning 4 evaluation dimensions and 15 sub-tasks. Each question is grounded in an explicit reasoning chain that averages 3.68 hops across a 15.1-minute temporal span. The dataset is built by a three-step semi-automated pipeline followed by a strict quality assurance process. Evaluation of multiple representative OmniLLMs on TraceAV-Bench reveals that the benchmark poses a persistent challenge across all models, with the strongest closed-source model (Gemini 3.1 Pro) reaching only 68.29% on general tasks, and the best open-source model (Ming-Flash-Omni-2.0) reaching 51.70%, leaving substantial headroom. Moreover, we find that robustness to multimodal hallucination is largely decoupled from general multimodal reasoning performance. We anticipate that TraceAV-Bench will stimulate further research toward OmniLLMs that can reason coherently and faithfully over long-form audio-visual content.
Authors: Xuyi Hu, Jin Lyu, Jiuming Liu, Yebin Liu, Silvia Zuffi, Liang An, Stefan Goetz
Abstract: 3D animal reconstruction in the wild remains challenging due to large species variation, frequent occlusions, and the prevalence of multi-animal scenes, while existing methods predominantly focus on single-animal settings. We present SAM 3D Animal, the first promptable framework for multi-animal 3D reconstruction from a single image. Built on the SMAL+ parametric animal model, our method jointly reconstructs multiple instances and supports flexible prompts in the form of keypoints and masks which enable more reliable disambiguation in crowded and occluded scenes. To train such a model, we further introduce Herd3D, a multi-animal 3D dataset containing over 5K images, designed to increase diversity in species, interactions, and occlusion patterns. Experiments on the Animal3D, APTv2, and Animal Kingdom datasets show that our framework achieves state-of-the-art results over both existing model-based and model-free methods, demonstrating a scalable and effective solution for prompt-driven animal 3D reconstruction in the wild.
Authors: Zhixin Cheng, Yujia Chen, Xujing Tao, Bohao Liao, Xiaotian Yin, Baoqun Yin, Tianzhu Zhang
Abstract: Image-to-point cloud registration is often challenged by viewpoint changes, cross-modal discrepancies, and repetitive textures, which induce scale ambiguity and consequently lead to erroneous correspondences. Recent detection-free methods alleviate this issue by leveraging multi-scale features and transformer-based interactions. However, they still suffer from attention drift across layers and intra-scale inconsistencies, hindering precise registration. Inspired by human behavior, we propose a ``Focus--Sweep'' paradigm and develop a Hierarchical Focus--Sweep Interaction Module within an SSM-based framework to enhance multi-level cross-modal feature association. In addition, we introduce a Dynamic Layer Allocation Strategy that adaptively determines the iteration depth to better exploit geometric constraints and improve matching robustness. Extensive experiments and ablations on two benchmarks, RGB-D Scenes V2 and 7-Scenes, demonstrate that our approach achieves state-of-the-art performance.
Authors: Jun Wang, Fengpeng Li, Hang Dong, Tianjin Huang, Wei Han
Abstract: Remote sensing lithology interpretation is fundamental to geological surveys, mineral exploration, and regional geological mapping. Unlike general land-cover recognition, lithology interpretation is a knowledge-intensive task that requires experts to infer rock types from various features, e.g., subtle visual, spectral, textural, geomorphological, and contextual cues, making reliable automated interpretation highly challenging. Geological knowledge-guided large multimodal models offer new opportunities, yet their evaluation remains constrained by the lack of benchmarks that capture lithological annotations, multi-level geological semantics, and expert-informed assessment. Here, we propose LithoBench, a multi-level benchmark for evaluating geological semantic understanding in remote sensing lithology interpretation. LithoBench contains 10,000 expert-annotated interpretation instances across 12 representative lithological categories, including 4,000 multiple-choice and 6,000 open-ended tasks organized into five cognitive levels: Identification and Description, Comparative Analysis, Mechanism Explanation, Practical Application, and Comprehensive Reasoning. We further develop an expert-in-the-loop, knowledge-grounded semi-automated construction pipeline, coupling multi sub-processes, e.g., structured geological image descriptions, to enhance geological validity and evaluation reliability. Experiments with multiple large vision-language models eveal substantial limitations in geological semantic understanding, particularly on higher-order explanation, application, and reasoning tasks.
Authors: Jaeyoung Choi, Hyeondong Kim, Yujin Kim, Daehee Park
Abstract: Forecasting future 3D hand pose sequences from egocentric video is essential for understanding human intention and enabling embodied applications such as AR/VR assistance and human-robot interaction. However, this task remains a highly challenging problem because egocentric hand motion is driven by complex human intent, exhibits highly dexterous articulations, and is observed under drastic viewpoint shifts induced by ego-motion. In this work, we introduce EggHand, a foundation-model-based framework for egocentric hand pose forecasting that unifies multimodal semantic reasoning with dynamic motion modeling. Our approach couples an action decoder from a Vision-Language-Action (VLA) model, which captures the structured temporal dynamics of hand motion, with an egocentric video-text encoder that provides viewpoint-aware contextual information learned from large-scale first-person video. Together, these components overcome the brittleness of generic visual encoders under ego-motion and enable joint reasoning over motion, context, and high-level intent-without relying on body pose or external tracking. Experiments on the EgoExo4D dataset show that EggHand sets a new state of the art in forecasting accuracy, remains robust under severe ego-motion, and further enables controllable prediction via language-based task prompts. Project page: https://jyoun9.github.io/EggHand
Authors: Berkehan \"Unal, Dierend Hauke, Fazlija Dren, Plachetka Christopher
Abstract: Over the last few years, research on autonomous systems has matured to such a degree that the field is increasingly well-positioned to translate research into practical, stakeholder-driven use cases across well-defined domains. However, for a wide-scale practical adoption of autonomous systems, adherence to safety regulations is crucial. Many regulations are influenced by the Operational Design Domain (ODD), which defines the specific conditions in which an autonomous agent can function. This is especially relevant for Automated Driving Systems (ADS), as a dependable perception of ODD elements is essential for safe implementation and auditing. Vision-language models (VLMs) integrate visual recognition and language reasoning, functioning without task-specific training data, which makes them suitable for adaptable ODD perception. To assess whether VLMs can function as zero-shot "ODD sensors" that adapt to evolving definitions, we contribute (i) an empirical study of zero-shot ODD classification and detection using four VLMs on a custom dataset and Mapillary Vistas, along with failure analyses; (ii) an ablation of zero-shot optimization strategies with a cost-performance overview; and (iii) a suite of reusable prompting templates with guidance for adaptation. Our findings indicate that definition-anchored chain-of-thought prompting with persona decomposition performs best, while other methods may result in reduced recall. Overall, our results pave the way for transparent and effective ODD-based perception in safety-critical applications.
Authors: Zijia Fu (School of Artificial Intelligence, Beijing Normal University, Beijing, China), Yuanfei Huang (School of Artificial Intelligence, Beijing Normal University, Beijing, China), Lizhi Wang (School of Artificial Intelligence, Beijing Normal University, Beijing, China), Hua Huang (School of Artificial Intelligence, Beijing Normal University, Beijing, China)
Abstract: Lens flares, caused by complex optical aberrations, severely degrade image quality especially in nighttime photography. Although recent restoration methods have made remarkable progress, most still rely on spatially uniform processing. They are failing to handle the region-dependent restoration demands of flare scenes, where saturated light sources should be preserved, flare artifacts removed, and background details recovered. To address this challenge, we propose DeflareMambav2, a prior-guided Mamba framework for lens flare removal. Specifically, we introduce a Flare Prior Network (FPN) to estimate flare priors and guide adaptive restoration. Besides, a novel radial serialization strategy breaks spatially homogeneous processing by performing flare-aware targeted sampling, and better supports long-range modeling in State Space Models (SSMs). Based on these priors, the backbone adopts a dual-level adaptive scheme. It explicitly preserves light-source regions to avoid over-processing, and applies curriculum-based restoration to the remaining contaminated areas while calibrating restoration intensity at the pixel level. Extensive experiments demonstrate that DeflareMambav2 achieves state-of-the-art performance with reduced parameter burden. Code is available at https://github.com/BNU-ERC-ITEA/DeflareMambav2.
Authors: Pei Zhang, Yunkai Liang, Kaiqiang Wang
Abstract: Underwater environments impose severe constraints on conventional imaging systems and demand solutions that balance high-quality sensing with strict resource efficiency. While emerging event cameras offer a promising alternative, their potential in aquatic scenarios remains largely unexplored. Through the lens of neuromorphic vision, this work pioneers the investigation of motion fields that serve as key media for agile underwater perception. Built upon spiking neural networks, we introduce a self-supervised framework to estimate per-pixel optical flow from asynchronous event streams, elegantly bypassing the long-standing bottleneck of underwater data scarcity. Extensive evaluations demonstrate that our method achieves competitive visual and quantitative results against leading techniques while operating with superior computational efficiency. By bridging neuromorphic sensing and aquatic intelligence, this work opens new frontiers for lightweight, real-time, and low-cost perception on resource-constrained underwater edge platforms.
Authors: Arka Koner, Chetan S. Naik, Lokesh Kurre, Vivek Raghavan, Barada P. Sabut, Tanusree Deb Barma, Anoop M. Namboodiri, Anil K. Jain
Abstract: Searching a multi-biometric database of a billion records for a country-level identity system requires pushing the limits of all aspects of a biometric system, including acquisition, preprocessing, feature extraction, accuracy, matching speed, presentation attack detection, and handling of special cases (e.g., missing finger digits). This is the first paper that gives insights into such a large-scale multimodal biometric search system, called Bharat ABIS, based on open-source architectures. The end-to-end pipeline of Bharat ABIS processes fingerprint, face and iris modalities through modality-specific stages of preprocessing (segmentation), quality assessment, presentation attack detection, and learning an embedding (feature extraction), producing a concatenated template of 13.5KB per person. We present a detailed analysis of the modalities and how they are integrated to create an efficient and effective solution for 1:N search (de-duplication). Evaluations on a demographically stratified gallery of 220 million identities, randomly sampled from 1.55 billion records in India's Aadhaar database, yield an FNIR of 0.3% at an FPIR of 0.5%, for adult probes (over 18 years). We also compare the performance of Bharat ABIS against three state-of-the-art COTS systems on a 20M gallery. Our system achieves a throughput of 100 searches per second on a gallery of 40M on a single server (8xNvidia H100 GPUs, 2TB RAM).
Authors: Ritul Jangir, Arkya Jyoti Bagchi, Aiman Farooq, Mangalton Okram, Saurabh Seetaram Korgaonkar, Deepak Mishra
Abstract: High-fidelity surgical video generation can greatly improve medical training and the development of AI, adapting these generative models for precise video editing remains a formidable challenge. Modifying surgical attributes, such as instrument tissue interactions or procedural phases is challenging due to the strict anatomical and temporal constraints. In this paper, we propose OphEdit, a novel training-free framework for the text-guided editing of ophthalmic surgical videos. Our approach leverages a deterministic second-order ODE inversion pipeline to capture Attention Value (V) tensors from the original video. By selectively injecting these stored tensors into the conditional Classifier-Free Guidance (CFG) branch during the denoising phase, OphEdit rigorously preserves the intricate anatomical geometry of the eye while seamlessly mapping text-driven semantic modifications onto the video stream. Clinical evaluations demonstrates that OphEdit effectively handles complex surgical transformations, such as instrument swaps and procedural variations, with superior structural fidelity and temporal consistency compared to natural-domain video editors. Our work represents the first application of training-free video editing in the ophthalmic surgical domain, offering a scalable solution for generating diverse, annotated medical datasets without the need for exhaustive manual recording or costly model fine-tuning. The code and prompts can be accessed at https://github.com/ophedit/OphEdit
Authors: Matthias Kahl, Zhaiyu Chen, Sudipan Saha, Mrinalini Kochupillai, Lukas Kondmann, Xiao Xiang Zhu
Abstract: Mining operations are of utmost importance to the economy of some nations. However, such operations result in land-use change, very high energy consumption, and negative impacts on the environment, including soil erosion and deforestation. The mining process can impact an area much larger than the mining site itself. Adding to the negative externalities linked to mining is the fact that, in addition to government-sanctioned legal mining operations, illegal mining is widespread, including in various countries of Africa. The ability to monitor remote mining site activities can be useful, e.g., for the detection of illegal artisanal mining activities and their environmental impacts. An important outcome of such monitoring could include a better understanding of the interrelationship between mine facility attributes (e.g., mining types, processing methods, commodities, etc.) and their impact on the natural environment. In this work, we present a data set that contains 150 Large Scale Mining (LSM) sites and 870km^2 annotated area of Artisanal Small-scale Mining (ASM) sites. The metadata includes nine eminent LSM sections and 27 mining site attributes for each LSM site. We also discuss the data set's possible contribution to the research community, social and environmental consequences, and researchers' responsibilities from an ethics perspective.
Authors: Hartmut H\"antze, Sarah de Boer, Myrthe Buser, Alessa Hering, Bram van Ginneken, Mathias Prokop, Jawed Nawabi, Sebastian Ziegelmayer, Lisa Adams, Keno Bressem
Abstract: The rapid proliferation of open-source medical foundation models (FMs) raises a practical question: how well do their pre-trained representations transfer to clinically relevant but data-scarce classification tasks? Particularly in CT-based renal lesion classification, a push toward greater generalizability would be meaningful, as the field is constrained by inherently limited training data. We addressed this through a benchmark of three medical FMs on this specific task. This six-class problem spans common entities like cysts and clear cell renal cell carcinoma, alongside rare subtypes. Using a frozen feature-probing protocol, we compared FM embeddings against a handcrafted radiomics classifier and a 3D ResNet-50 trained from scratch. Models were trained on a composite dataset of 2,854 lesions and evaluated on an external test set of 234 lesions from The Cancer Imaging Archive. Our results reveal two key findings. First, FM performance (AUC 0.70-0.77) matched the from-scratch ResNet (AUC 0.72) while drastically reducing hardware demand, requiring only seconds on a CPU after feature extraction. However, the conventional radiomics baseline significantly outperformed all deep learning approaches, achieving an AUC of 0.88 (all p $\leq$ 0.002). This suggests that current generalist FM embeddings do not yet capture the fine-grained texture and shape heterogeneity driving histological subtype discrimination. Despite their potential in data-scarce settings, medical FMs did not surpass established models for renal lesion stratification, leaving radiomics as the current state-of-the-art.
Authors: Yingfeng Wang, Yuxuan Xiao, Shengcai Liao
Abstract: Many vision applications require identity consistency beyond strict biometric recognition, especially under non-frontal views or when facial cues are missing. However, conventional face recognition models enforce intra-identity invariance, collapsing appearance variations such as hairstyle or styling changes into a single representation, limiting their use in appearance-sensitive scenarios. To address this limitation, we introduce Head Similarity, a new formulation that extends identity-centric recognition to structured whole-head similarity modeling. Our approach explicitly captures intra-identity appearance variation and enforces hierarchical similarity ordering across identity and appearance states, enabling meaningful comparison even under occlusion or rear-view conditions. We construct a large-scale benchmark from long-form videos with weakly-supervised appearance states, covering diverse poses, occlusions, and temporal changes. As a first step, we develop a simple yet effective framework that jointly models identity discrimination and appearance-sensitive similarity through hierarchical supervision and identity-aware distillation. Experiments show that conventional face recognition models fail to capture appearance-dependent similarity, while our approach demonstrates the feasibility of structured whole-head similarity modeling.
Authors: Xuanshuo Fu, Lei Kang, Javier Vazquez-Corral
Abstract: Poor lighting conditions significantly impact image quality, posing substantial challenges for image editing and visualization. Many existing enhancement methods aim at proposing complex models while neglecting the intrinsic information contained within low-light images. In this work, we propose the Self-Information Mining (SIMI) network, an innovative unsupervised framework that decomposes low-light images into multiple components based on bit-plane decomposition. Our approach allows mining intrinsic information without relying on external data. This not only accelerates model convergence but also improves performance and reduces computational overhead. The unsupervised nature of our method facilitates real-world applicability. Experiments conducted on standard benchmarks demonstrate that SIMI achieves state-of-the-art performance.
Authors: Niklas Vaara, Lam Huynh, Pekka Sangi, Miguel Bordallo L\'opez, Janne Heikkil\"a
Abstract: Explicit neural representations such as 3D Gaussian Splatting (3DGS) enable high-fidelity and real-time novel view synthesis, yet optimize for alpha-composited optical appearance rather than ray-intersectable geometry. In contrast, radio-frequency (RF) digital twins require deterministic multi-bounce paths, where the geometry dictates trajectories and their associated attenuation and delay. We introduce a framework enabling differentiable RF propagation simulation directly within visually reconstructed neural scenes, allowing point-to-point path computation between arbitrary 3D locations while preserving high-quality visual rendering. Unlike conventional RF simulation pipelines that rely on manually constructed meshes, we embed Gaussian primitives into a hardware-accelerated ray tracing structure as the underlying spatial representation. By extracting physically meaningful channel impulse responses from visual-only reconstructions, we provide cross-modal evidence that neural reconstructions can serve as unified spatial representations for both electromagnetic propagation simulation and photorealistic view synthesis.
Authors: Amy Rafferty, Rishi Ramaesh, Ajitha Rajan
Abstract: Concept Bottleneck Models (CBMs) in medical imaging aim to improve model interpretability by predicting intermediate clinical concepts before final diagnoses. However, most existing CBMs treat concepts as discriminative predictors of pathology labels, without explicitly modelling the underlying clinical generative process where diseases produce observable radiographic findings. We propose XpertCausal, a radiologist-guided causal CBM for chest X-ray interpretation which models pathology-to-concept relationships using a probabilistic noisy-OR framework. This generative model is then inverted via Bayesian inference to estimate pathology probabilities from predicted concepts. Radiologist-curated concept-pathology associations are used to constrain model structure to radiologist-defined clinically plausible reasoning pathways. We evaluate XpertCausal on MIMIC-CXR across pathology classification performance, calibration, explanation quality, and alignment with radiologist-defined reasoning pathways. Compared with both a non-causal CBM baseline and a causal ablation with unconstrained learned associations, XpertCausal achieves improved AUROC, calibration, and clinically relevant explanation quality, while learning concept-pathology relationships that more closely align with expert knowledge. These results demonstrate that incorporating clinically motivated causal structure and expert domain knowledge into CBMs can lead to more accurate, interpretable, and clinically aligned models for CXR interpretation.
Authors: Caterina Gallegati, Monica Bianchini, Franco Scarselli, Vittorio Murino, Barbara Toniella Corradini
Abstract: As generative models achieve unprecedented visual quality, the gold standard for image evaluation remains traditional feature-distribution metrics (e.g., FID). However, these metrics are provably hindered by the closed-vocabulary bottleneck of outdated features and the assumptive bias of rigid parametric formulations. Recent alternatives exploit modern backbones to solve the feature bottleneck, yet continue to suffer from parametric limitations. To close this gap, we introduce APEX (Assumption-free Projection-based Embedding eXamination), a novel evaluation framework leveraging the Sliced Wasserstein Distance as a mathematically grounded, assumption-free similarity measure. APEX inherits effective scalability to high-dimensional spaces, as we prove with theoretical and empirical evidences. Moreover, APEX is embedding-agnostic and uses two open-vocabulary foundation models, CLIP and DINOv2, as feature extractors. Benchmarking APEX against established baselines reveals superior robustness to visual degradations. Additionally, we show that APEX metrics exhibit intra- and cross-dataset stability, ensuring highly stable evaluations on out-of-domain datasets.
Authors: Jiesong Lian, Zixiang Zhou, Ruizhe Zhong, Yuan Zhou, Qinglin Lu, Rui Wang, Long Hu, Yixue Hao, Baoru Huang
Abstract: Recent video diffusion models (VDMs) synthesize visually convincing clips, yet still drop entities, mis-bind attributes, and weaken the interactions specified in the prompt. Representation-alignment objectives such as VideoREPA and MoAlign improve fine-grained text following by distilling spatio-temporal token relations from a frozen visual foundation model, but their pairwise supervision budget is allocated by visual or motion cues rather than by how relevant each pair is to the prompt. We present SARA, Semantically Adaptive Relational Alignment, which keeps token-relation distillation (TRD) on a frozen VFM target and adds a text-conditioned saliency that decides which token pairs carry supervision. A lightweight Stage 1 aligner is trained with per-entity SAM 3.1 mask supervision and an InfoNCE regulariser, and its continuous saliency is fused into TRD through a pair-routing operator that assigns each token pair a weight whenever either of its two endpoints is salient, thereby routing supervision toward subject-subject and subject-background pairs and away from background-background ones. In the Wan2.2 continual-training setting, SARA improves both text alignment and motion quality over SFT, VideoREPA, and MoAlign on a 13-dimension VLM rubric, on the public VBench benchmarks, and in a blind user study.
Authors: Dimitrios Mallis, Marco Wang, Ahmet Serdar Karadeniz, Elisa Ricci, Anis Kacem, Djamila Aouada
Abstract: Text-to-CAD has recently emerged as an important task with the potential to substantially accelerate design workflows. Despite its significance, there has been surprisingly little work on Text-to-CAD evaluation, and assessing CAD model generation performance remains a considerable challenge. In this work, we introduce a new evaluation perspective for Text-to-CAD based on automated testing. We propose CADTestBench, the first test-based benchmark for Text-to-CAD, based on CADTests, executable software tests that verify whether a generated CAD model satisfies the geometric and topological requirements of the input prompt. Using CADTestBench, we conduct comprehensive benchmarking of recent Text-to-CAD methods and further demonstrate that CADTests can also guide CAD model generation, yielding simple baselines that surpass performance of current methods. CADTestBench code and data are available at GitHub and Hugging Face dataset.
Authors: Thomas Gorges, Janne van der Loop, Lukas H\"uttner, Linda-Sophie Schneider, Fei Wu, Mathias Seuret, Vincent Christlein
Abstract: This paper presents CircleID, a large-scale ICDAR 2026 competition on writer identification and pen classification from scanned hand-drawn circles. The primary objective is to investigate how biometric writer characteristics and physical pen features naturally entangle within minimal, static traces. CircleID comprises two distinct tasks: (1) open-set writer identification, requiring models to recognize known writers while explicitly rejecting unknown ones, and (2) cross-writer pen classification, evaluated across both seen and unseen writers. Participants were provided with a new, controlled dataset of 46,155 tightly cropped circle images, digitized at 400 DPI and annotated for writer identity and pen type. The dataset comprises samples from 50 known and 16 unknown writers using eight different pens. Hosted on Kaggle as two separate tracks with public and private leaderboards, the competition provided participants with a ResNet baseline. In total, 389 teams (436 participants) made 3,185 submissions for the pen classification task, and 113 teams (141 participants) made 1,737 submissions for the writer identification track. The best-performing private leaderboard submissions achieved a Top-1 accuracy of 64.801% for writer identification and 92.726% for pen classification. This paper details the dataset, evaluates the winning methodologies, and analyzes the impact of out-of-distribution writers on model generalization and feature disentanglement. In this large-scale competition, CircleID establishes a new baseline for minimal-trace analysis.
Authors: Brown Ebouky, Gabriele Carrino, Niccolo Avogaro, Christoph Studer, Andrea Bartezzaghi, Mattia Rigotti
Abstract: Human visual reasoning is governed by active vision, a process where metacognitive control drives top-down goal-directed attention, dynamically routing foveal focus toward task-relevant details while maintaining peripheral awareness of the global scene. In contrast, modern Vision-Language Models (VLMs) process visual information passively, relying on the static accumulation of massive token contexts that dilute spatial reasoning and induce linguistic hallucinations. Here we propose the following paradigm shift: GazeVLM, a multimodal architecture that internalizes this metacognitive oversight over its deployment of attention resources directly into the reasoning loop. By empowering the VLM to autonomously generate gaze tokens ($\texttt{
Authors: Boyang Dai, Chaoqi Chen, Yizhou Yu
Abstract: Out-of-distribution (OOD) detection is crucial for ensuring the reliability of deep learning models. Existing methods mostly focus on regular entangled representations to discriminate in-distribution (ID) and OOD data, neglecting the rich contextual information within images. This issue is particularly challenging for detecting near-OOD, as models with simplicity bias struggle to learn discriminative features in disentangled representations. The human visual system can use the co-occurrence of objects in the natural environment to facilitate scene understanding. Inspired by this, we propose an Object-Centric OOD detection framework that learns to capture Object CO-occurrence (OCO) patterns within images. The proposed method introduces a new OOD detection paradigm that understands object co-occurrence within an image by predicting disentangled representations for the test sample, then adaptively divides patterns into three scenarios based on object co-occurrence patterns observed in ID training data, and finally performs OOD detection in a divide-and-conquer manner. By doing so, OCO can distinguish near-OOD by considering the semantic contextual relationships present in their images, avoiding the tendency to focus solely on simple, easily learnable regions. We evaluate OCO through experiments across challenging and full-spectrum OOD settings, demonstrating competitive results and confirming its ability to address both semantic and covariate shifts. Code is released at https://github.com/Michael-McQueen/OCO.
Authors: Andreas Caduff (Competence Center for Intelligent Sensors and Networks, Lucerne University of Applied Science and Art), Klaus Zahn (Competence Center for Intelligent Sensors and Networks, Lucerne University of Applied Science and Art), Jonas Hofstetter (Competence Center for Intelligent Sensors and Networks, Lucerne University of Applied Science and Art), Martin Rechsteiner (Competence Center for Intelligent Sensors and Networks, Lucerne University of Applied Science and Art), Patrick Flaig (SICK AG)
Abstract: In the area of Intelligent Transportation Systems (ITS), fine-grained vehicle classification systems play an essential role. Recently, the authors have presented a novel vision-based classification approach in which standard end-to-end Convolutional Neural Networks (CNNs) have been decomposed into 1) a CNN-based detector for semantically strong vehicle parts, followed by 2) feature construction and 3) final classification by a decision tree. In contrast to conventional CNNs, this allows both easy extensibility to new vehicle categories - without the need to fully retrain the part detector - and an important step towards the interpretability of the model, removing partially the black-box nature inherent to CNNs. Here we present an important extension of this approach that now incorporates spatial awareness of the vehicle parts: while the feature construction 2) of the previous approach used a binary decision for each feature (present vs. absent), now a full spatial probability map is constructed to condition the presence of each individual part with respect to a given vehicle category. The classification is performed using a softmax regression approach for the overall vehicle probabilities. This method shows a considerably improved robustness against false (part-)detections, a point that is crucial for practical application. Comparative analyses with a state-of-the-art end-to-end CNN indicate that our part-based methods achieve comparable accuracy, effectively challenging the presumed trade-off between accuracy and explainability. This research represents a significant advance in vehicle classification for ITS and forms the basis for systems that combine high accuracy with intuitive interpretability.
Authors: Peilin Xiong, Honghui Yuan, Junwen Chen, Keiji Yanai
Abstract: Coarse-mask local image editing asks a model to modify a user-indicated region while preserving the surrounding scene. In practice, however, rough masks often become unintended shape priors: instead of serving as flexible edit support, the mask can pull the generated object toward its accidental boundary. We study this failure as mask-shape bias and frame the task through a Two-Zone Constraint, where the background should remain stable while the editable region should follow the instruction without being forced to inherit the mask contour. BRIDGE addresses this setting by keeping masks outside the DiT backbone for support construction and blending, avoiding DiT-internal mask injection and copied control branches. It uses BridgePath generation, where a Main Path preserves background context and a Subject Path generates editable content from independent noise. Motivated by a diagnostic Qwen-Image experiment showing that positional embeddings and attention connectivity regulate which image context visual tokens reuse, BRIDGE introduces a learnable Discrete Geometric Gate for token-level positional-embedding routing. This gate lets subject tokens borrow background-anchored coordinates near fusion regions or keep subject-centric coordinates for geometric freedom. We evaluate BRIDGE on BRIDGE-Bench, MagicBrush, and ICE-Bench. On BRIDGE-Bench, BRIDGE improves Local SigLIP2-T from 0.262 with FLUX.1-Fill and 0.390 with ACE++ to 0.503, with parallel gains in local DINO and DreamSim. Zero-shot results on MagicBrush and ICE-Bench further indicate competitive alignment and source preservation beyond the curated benchmark, while the added routing module remains compact at 13.31M parameters compared with ControlNet-style copied branches.
Authors: Lang Zhang, JinYi Yoon, Matthew Corbett, Abhijit Sarkar, Bo Ji
Abstract: Driver cognitive distraction is a major cause of road collisions and remains difficult to detect. Unlike manual or visual distraction, cognitive distraction is diverted by thoughts unrelated to driving, even when the driver appears visually attentive and exhibits no explicit physical movements. In this work, we propose EyeCue, a gaze-empowered egocentric video understanding framework, to detect driver cognitive distraction. A key insight is that cognitive distraction manifests in the interaction between eye gaze and visual context. To capture this interaction, EyeCue integrates eye gaze with egocentric video to enable context-aware modeling of the driver's attention over time. Furthermore, to tackle the limited scale and diversity of existing datasets, we introduce CogDrive, a comprehensive multi-scenario dataset that augments four existing driving datasets with cognitive distraction annotations. Through extensive evaluations on CogDrive, we show that EyeCue achieves the highest accuracy of 74.38%, outperforming 11 baselines from 6 model families by over 7%. Notably, EyeCue can achieve an accuracy of over 70% across various driving scenarios (different road types, times of day, and weather conditions) with strong generalizability. These results highlight the importance of modeling gaze-context interactions and the effectiveness of cross-modal interaction modeling for multimodal cognitive distraction detection. Our codes and CogDrive dataset resources are available at https://github.com/langzhang2000/EyeCue.
Authors: Yue Yu, Jiayu Wang, Jiajia Shi, Jingjing Chen, Yu-Gang Jiang
Abstract: Makeup transfer aims to apply the makeup style of a reference portrait to a source portrait while preserving identity and background. Early methods formulate this task as unsupervised image-to-image translation, relying on surrogate objectives and often yielding limited performance. Recent diffusion- and flow-based approaches instead exploit synthetic data for supervised training, leading to significant improvements. However, these methods still face two critical challenges: synthetic supervision frequently fails to faithfully preserve identity, and the domain gap between synthetic and real data limits generalization, resulting in degraded performance in complex real-world scenarios. To address these issues, this paper first proposes ConsistentBeauty, a novel data curation pipeline that ensures makeup fidelity and strict identity consistency within the synthesized data. Second, we propose RealBeauty, a synthetic-to-real post-training framework. Beyond supervised learning on curated synthetic data, we further adapt the model to real-world scenarios through reinforcement learning and design novel verifiable rewards tailored to the makeup transfer task. It allows the model to further benefit from real makeup patterns beyond synthetic supervision. In addition, we establish a new diverse benchmark for makeup transfer, covering a wide range of skin tones, ages, genders, poses, and makeup styles, thereby enabling a more comprehensive evaluation of model performance under diverse real-world conditions. Extensive experiments show that our method achieves state-of-the-art performance on multiple benchmarks and demonstrates clear advantages in identity preservation and performance on complex real-world cases.
Authors: Yuancheng Wei, Linli Yao, Lei Li, Haojie Zhang, Hao Zhou, Fandong Meng, Xu Sun
Abstract: Multimodal reward models have advanced substantially in text and image domains, yet progress in video understanding reward modeling remains severely limited by the lack of robust evaluation benchmarks and high-quality preference data. To address this, we propose a unified framework spanning benchmark design, data construction, and reward model training. We introduce Video Understanding Reward Bench (VURB), a benchmark featuring 2,100 preference pairs with long chain-of-thought reasoning traces (averaging 1,143 tokens) and majority voting evaluation across general, long, and reasoning-oriented video tasks. We further construct Video Understanding Preference Dataset (VUP-35K) via a fully automated pipeline, providing large-scale high-quality supervision for video reward training. Building on the data, we train VideoDRM and VideoGRM, a discriminative and a generative reward model, both achieving state-of-the-art performance on VURB and VideoRewardBench. Further analysis confirms that VUP-35K enhances both reward performance and model reasoning capability, while VideoDRM and VideoGRM yield significant gains under best-of-$N$ test-time scaling.
Authors: Hang Wu, Sherin Mary Mathews, Yujun Cai, Ming-Hsuan Yang, Yiwei Wang
Abstract: Online streaming video understanding requires models to process continuous visual inputs and respond to user queries in real time, where the unbounded stream and unpredictable query timing turn memory management into a central challenge. Existing methods typically compress visual tokens via visual similarity heuristics, or augment compression with KV-cache-level retrieval. However, compression decisions rarely incorporate semantic signals, and retrieval is often added after compression is finalized, making the two stages hard to coordinate. We present SAVEMem, a training-free dual-stage framework that brings semantic awareness into memory generation and lets the retrieval scope adapt per query. In Stage~1, SAVEMem builds a three-tier streaming memory online under a constant memory budget. A fixed pseudo-question bank provides a lightweight semantic prior, so that long-term retention is shaped by semantic salience rather than visual similarity alone. In Stage~2, SAVEMem performs query-aware retrieval over this memory. An anchor-conditioned recency gate adapts the retrieval scope from short-term to mid- and long-term memory based on whether the query targets the present or the distant past. Within this scope, late interaction between query and memory tokens selects candidate frames for answering. Applied to Qwen2.5-VL without training, SAVEMem improves the OVO-Bench overall score from 52.27 to 62.69 and yields consistent gains on StreamingBench and ODV-Bench, while reducing peak GPU memory by 48\% at 128 frames over the backbone.
Authors: Yulong Chen, Xiaoyun Dong, Haoyu Zhang, Zongxian Yang, Lewei Xie, Xinke Li, Yifan Zhang, Kai Wang, Jianping Wang
Abstract: Reconstructing dynamic scenes from Vehicle-to-Infrastructure Cooperative Autonomous Driving (VICAD) data is fundamentally complicated by temporal asynchrony: vehicle and infrastructure cameras operate on independent clocks, capturing the same dynamic agent such as cars and pedestrians at different physical times. Existing Gaussian Scene Graph methods implicitly assume synchronized observations and assign a single pose per agent per frame, which is an assumption that breaks in cooperative settings, where the resulting gradient conflicts cause severe ghosting on dynamic agents. We identify this as a representation-level failure, not an optimization artifact: we prove that any single-timeline formulation incurs an irreducible photometric loss scaling quadratically with agent velocity and cross-source time offset. To resolve this, we propose Dust (DecoUpled Spatio-Temporal) Gaussian Scene Graph for 4D Cooperative Driving Reconstruction. DUST Gaussian Scene Graph shares a canonical Gaussian set per agent for appearance consistency, while maintaining decouple pose trajectories aligned to each source's true capture timestamps. We prove that this decoupling enables the pose-gradient kernel block-diagonal, eliminating cross-source interference entirely. To make Dust practical, we further introduce a static anchor-based pose correction pipeline that corrects spatio misalignment between vehicle and infrastructure annotations, and a pose-regularized joint optimization scheme that prevents trajectory jitter and drift during early training. On 26 sequences from V2X-Seq, DUST achieves state-of-the-art performance, improving dynamic-area PSNR by 3.2 dB over the strongest baseline and reducing Fr\'echet Video Distance by 37.7%, with keeping robustness under larger temporal asynchrony. Code is available at https://anonymous.4open.science/r/DUST-6A55.
Authors: Zhengrong Yue, Taihang Hu, Mengting Chen, Haiyu Zhang, Zihao Pan, Tao Liu, Zikang Wang, Jinsong Lan, Xiaoyong Zhu, Bo Zheng, Yali Wang
Abstract: Tokenizers are a crucial component of latent diffusion models, as they define the latent space in which diffusion models operate. However, existing tokenizers are primarily designed to improve reconstruction fidelity or inherit pretrained representations, leaving unclear what kind of latent space is truly friendly for generative modeling. In this paper, we study this question from the perspective of latent manifold organization. By constructing controlled tokenizer variants, we identify three key properties of a diffusion-friendly latent manifold: coherent spatial structure, local manifold continuity, and global manifold semantics. We find that these properties are more consistent with downstream generation quality than reconstruction fidelity. Motivated by this finding, we propose the Prior-Aligned AutoEncoder (PAE), which explicitly shapes the latent manifold instead of leaving diffusion-friendly manifold to emerge indirectly from reconstruction or inheritance. Specifically, PAE leverages refined priors derived from VFMs and perturbation-based regularization to turn spatial structure, local continuity, and global semantics into explicit training objectives. On ImageNet 256x256, PAE improves both training efficiency and generation quality over existing tokenizers, reaching performance comparable to RAE with up to 13x faster convergence under the same training setup and achieving a new state-of-the-art gFID of 1.03. These results highlight the importance of organizing the latent manifold for latent diffusion models.
Authors: Hanqi Jiang, Junhao Chen, Yi Pan, Lifeng Chen, Weihang You, Haozhen Gong, Ruiyu Yan, Jinglei Lv, Lin Zhao, Hui Ren, Quanzheng Li, Tianming Liu, Xiang Li
Abstract: Medical vision--language models (VLMs) are usually evaluated on intact image--question pairs, but trustworthy clinical use requires a stronger property: a model must recognise when the evidential basis for an answer has failed. We study this through silent failures under perturbed evidence, where a vision-required medical question is paired with a false premise, wording perturbation, knowledge-only rewrite, or ROI-corrupted image, yet the model returns a fluent non-refusal answer. We introduce medvigil, a 300-case evaluation suite drawn from four public medical VQA sources, supervised end to end by four board-certified radiologists: every gold answer, refusal option, candidate-answer set, paraphrase, false-premise trap, ROI box, and clinical risk tier is clinician-authored. Two attending radiologists annotate every case in parallel, a senior radiologist consolidates the released manifest, and a separate fourth radiologist independent of construction answers every probe to provide the human reference baseline. The release contains 2{,}556 MCQ probes, 240 counterfactual triplets, physician-adjudicated risk-tier and answerability flags, ROI boxes, and a paired open-ended variant. We report seven correctness-conditioned audit metrics that summarise into the medvigil Composite Score (MCS), and audit 16 vision-capable models plus two text-only baselines. The independent radiologist scores MCS 83.3 at silent-failure rate 5.8%, leaving a 14.1-point composite headroom above the strongest audited model (Claude Opus 4.7 at 69.2). The benchmark and evaluation harness are publicly released.
Authors: Zuojin Tang, Shengchao Yuan, Xiaoxin Bai, Zhiyuan Jin, De Ma, Gang Pan, Bin Liu
Abstract: Vision-language-action (VLA) models increasingly rely on auxiliary world modules to plan over long horizons, yet how such modules should be parameterized on top of a pretrained VLA remains an open design question. Existing world-model-augmented VLAs typically pass the per-frame visual stream into the world module at high visual bandwidth and treat its rollout as a side product of action prediction; under a constrained adaptation budget on a frozen backbone, this leaves both the per-frame representation and the latent action coupling under-examined. We introduce OneWM-VLA, which compresses each view into a single semantic token per frame through an Adaptive Attention Pooling, and produces the resulting latent stream and the action trajectory under a single flow-matching objective rather than connecting them through a separate decoder. Empirically, we find that per-frame visual bandwidth can be reduced to a single token without compromising long-horizon performance under our setup. Trained with 14.71M LoRA parameters on a $\pi_0$ (2B) backbone, OneWM-VLA improves the average success rate from 47.9% to 61.3% on MetaWorld~MT50, reaches 95.6% on LIBERO-Long (vs.85.2% for $\pi_0$), and reaches 60.0% on the long-horizon deformable task Fold Cloth on a real Piper arm (vs.20.0% for $\pi_0$).
Authors: Jiacheng Chen, Songze Li, Han Fu, Baoquan Zhao, Wei Liu, Yanyan Liang, Li Qing, Xudong Mao
Abstract: Exemplar-based image editing applies a transformation defined by a source-target image pair to a new query image. Existing methods rely on a pair-of-pairs supervision paradigm, requiring two image pairs sharing the same edit semantics to learn the target transformation. This constraint makes training data difficult to curate at scale and limits generalization across diverse edit types. We propose Delta-Adapter, a method that learns transferable editing semantics under single-pair supervision, requiring no textual guidance. Rather than directly exposing the exemplar pair to the model, we leverage a pre-trained vision encoder to extract a semantic delta that encodes the visual transformation between the two images. This semantic delta is injected into a pre-trained image editing model via a Perceiver-based adapter. Since the target image is never directly visible to the model, it can serve as the prediction target, enabling single-pair supervision without requiring additional exemplar pairs. This formulation allows us to leverage existing large-scale editing datasets for training. To further promote faithful transformation transfer, we introduce a semantic delta consistency loss that aligns the semantic change of the generated output with the ground-truth semantic delta extracted from the exemplar pair. Extensive experiments demonstrate that Delta-Adapter consistently improves both editing accuracy and content consistency over four strong baselines on seen editing tasks, while also generalizing more effectively to unseen editing tasks. Code will be available at https://delta-adapter.github.io.
Authors: Marwane Hariat, Antoine Manzanera, David Filliat
Abstract: We present CoopNet, an approach that improves the cooperation of co-trained networks by dynamically adapting the apportionment of gradient, to ensure equitable learning progress. It is applied to motion-aware self-supervised prediction of depth maps, by introducing a new hybrid loss, based on a distribution model of photo-metric reconstruction errors made by, on the one hand the depth + odometry paired networks, and on the other hand the optical flow network. This model essentially assumes that the pixels from moving objects (that must be discarded for training depth and odometry), correspond to those where the two reconstructions strongly disagree. We justify this model by theoretical considerations and experimental evidences. A comparative evaluation on KITTI and CityScapes datasets shows that CoopNet improves or is comparable to the state-of-the-art in depth, odometry and optical flow predictions.
Authors: Vicent Caselles-Ballester, Eloy Mart\'inez-Heras, Giuseppe Pontillo, Zoe Mendelsohn, Elena M. Marr\'on, Juan Luis Garc\'ia Fern\'andez, Laia Subirats, Jon Stutters, Jeremy Chataway, Frederik Barkhof, Sara Llufriu, Ferran Prados
Abstract: Multiple sclerosis (MS) expresses substantial clinical and radiological heterogeneity, which poses significant challenges for automatic lesion segmentation. The current deep learning-based SOTA is highly susceptible to changes in both distribution, e.g., changes in scanner; as well as the structure of inputs, evident in the current divide between cross-sectional and longitudinal approaches. We introduce TimeLesSeg, a unified contrast-agnostic framework designed to segment MS lesions regardless of the presence of a temporal dimension in its inputs, with a single convolutional neural network. Our approach models pathological priors through lesion masks, which are processed together with the current scan. Cross-sectional processing is enabled by exposing the model to training cases where no prior information is available, which are modeled with an empty mask, allowing it to operate seamlessly in both scenarios. To overcome the scarcity and inconsistency of longitudinal datasets, we propose a novel generative pipeline in which patterns of lesion evolution are simulated by stochastically deforming each individual lesion with morphological operations, producing realistic prior timepoints. In parallel, we achieve contrast agnosticism through Gaussian mixture model-based domain randomization, enabling the network to experience a wide spectrum of intensity profiles. Results on three publicly available and two in-house datasets show that TimeLesSeg outperforms the contrast-agnostic state of the art on single-modality inputs across overlap- and distance-based metrics. In longitudinal processing, our method outperforms SAMSEG, and captures lesion load dynamics more accurately than both the former and LST-AI. All source code related to the development of TimeLesSeg is available at https://github.com/NeuroADaS-Lab/TimeLesSeg.
Authors: Zhengrui Xiang, Jiaqi Wu, Fupeng Sun, Heliang Zheng, Yingzhen Li
Abstract: We introduce Discrete Voxel Diffusion (DVD), a discrete diffusion framework to generate, assess, and edit sparse voxels for SLat (Structured LATent) based 3D generative pipelines. Although discrete diffusion has not generally displaced continuous diffusion in image-like generation, we show that it can be an effective first-stage prior for sparse voxel scaffolds. By treating voxel occupancy as a native discrete variable, DVD avoids continuous-to-discrete thresholding and provides a simple framework for voxel generation, uncertainty estimation, and editing. Beyond quality gains, DVD provides more interpretable generation dynamics through explicit categorical modeling. Furthermore, we leverage the predictive entropy as a robust uncertainty metric to identify ambiguous voxel regions and complicated samples, facilitating tasks such as data filtering and quality assessment. Finally, we propose a lightweight fine-tuning strategy using block-structured perturbation patterns. This approach empowers the model to inpaint and edit voxels within a single sampling round, requiring negligible auxiliary computation and no additional model evaluations.
Authors: Arani Roy, Shristi Das Biswas, Kaushik Roy
Abstract: Text-to-image diffusion models can generate visually stunning images, yet, controlling what appears and how it appears, remains surprisingly difficult, especially when operating solely within the constraints of the text-conditioning space. For example, changing a subject or adjusting an attribute often leads to unintended side effects, such as altered backgrounds or distorted details. This is because most existing text-based control methods treat the embedding space as Euclidean and apply simple linear transformations, which do not reflect how semantic concepts are actually organized. In this work, we take a step back and ask: what is the true geometry of these embeddings? We find that text encoder representations lie on a hypersphere, where concepts are not linear directions but structured, anisotropic distributions better captured by Kent distributions. Building on this insight, we propose HEART, a training-free framework that performs Kent-aware geodesic transformations directly on the hypersphere. By respecting the underlying geometry, HEART enables intuitive and precise edits, such as consistent subject replacement and fine-grained attribute control, while preserving the original scene. Importantly, HEART requires no finetuning, inversion, or optimization, and generalizes across diffusion model architectures. Our results show that a simple shift in perspective, from linear to spherical, can unlock fast, and controllable image generation.
Authors: Qiwei Wang, Zhongyao Tuo, Xianghui Ze, Yujiao Shi
Abstract: Cross-view localization classically asks: where does this ground image lie on the satellite tile? Existing methods are typically limited to 3-DoF estimates -- an $(x,y)$ position and a yaw angle -- because nadir satellite imagery provides no direct cues for roll, pitch, or altitude, forcing a reliance on planar-motion and zero-tilt assumptions. These assumptions break on real terrain with slopes, ramps, and tilted camera mounts. To overcome this, we introduce a single UAV image as an intermediate viewpoint: it reveals the 3D structure invisible from nadir, supplies the cues for roll, pitch, and altitude that the satellite alone cannot provide, and needs only spatial overlap with the ground camera -- no known relative pose is required. Building on this insight, we propose **Cross3R**, a flexible feed-forward model that ingests a satellite tile together with a UAV image, a ground image, or both, and, in a single forward pass, recovers a cross-view 3D point cloud, the 6-DoF poses of every input camera, and the on-tile $(x,y)$ position and yaw of each perspective camera. For training and evaluation, we also construct **CrossGeo**, a 278K-image tri-view dataset spanning 85 scenes across every continent except Antarctica. On CrossGeo, Cross3R consistently outperforms feed-forward 3D baselines in point-cloud reconstruction, 6-DoF camera-pose estimation, and cross-view localization. On KITTI, it outperforms dedicated cross-view methods trained on KITTI on most metrics, despite having no KITTI training itself.
Authors: Praroop Chanda, Suryansh Kumar
Abstract: Recent progress in dense optical flow has been driven by increasingly complex architectures and multi-step refinement for test-time scaling. While these approaches achieve strong benchmark performance, they also require substantial computation during inference. This raises a fundamental question: Is scaling test-time computation the only way to improve dense optical flow accuracy? We argue that it is not. Instead, powerful visual semantic and geometric priors encoded in modern foundation models can reduce, if not overcome, the need for computationally expensive iterative refinement at test-time. In this paper, we present a framework that estimates dense optical flow in a single forward pass, leveraging pretrained foundation representations, while avoiding iterative refinement and additional inference-time computation, thus offering an alternative to test-time scaling. Our method extracts visual semantic features from a frozen DINO-v2 backbone and combines them with geometric cues from a monocular depth foundation model. We fuse these complementary priors into a unified representation and apply a global matching formulation to estimate dense correspondences without recurrent updates or test-time optimization. Despite avoiding iterative refinement, our approach achieves strong cross-dataset generalization across challenging benchmarks. On Sintel Final, we obtain 2.81 EPE without refinement, significantly improving over state-of-the-art (SOTA) SEA-RAFT under comparable training conditions and outperforming RAFT, GMFlow (without refinement), and recent FlowSeek in the same setting. These results suggest that strong foundation priors can substitute for test-time scaling, offering a computationally efficient alternative to refinement-heavy pipelines.
Authors: Chao Huang, Penfei Wei, Wei Wang, Jie Wen, Zhihua Wang, Li Shen, Wenqi Ren, Xiaochun Cao
Abstract: Video anomaly detection (VAD) aims to automatically identify events that deviate from normal patterns in untrimmed surveillance videos. Existing methods universally depend on large-scale annotations or task-specific training procedures, severely limiting their rapid deployment to novel scenes. We observe that intermediate-layer features of pre-trained multimodal large language models (MLLMs) already encode rich anomaly semantics, yet existing approaches rely on the language output pathway and fail to exploit the geometric discriminability latent in these representations. Based on this finding, we propose SphereVAD, a fully training-free, zero-shot VAD framework that recasts anomaly discrimination as von Mises-Fisher (vMF) likelihood-ratio geodesic inference on the unit hypersphere, unleashing latent discriminability through principled geometric reasoning rather than learning new representations. Specifically, SphereVAD first applies Frechet mean centering to unfold feature distributions and eliminate domain biases, then employs Holistic Scene Attention (HSA) to reinforce feature consistency using cross-video priors, and finally performs vMF-guided Spherical Geodesic Pulling (SGP) to align ambiguous segments with directional prototypes on the spherical manifold. This training-free pipeline requires only minimal synthetic images for calibration. SphereVAD establishes new state-of-the-art results among training-free approaches on three major benchmarks and remains competitive with fully supervised baselines. Code will be available upon acceptance.
Authors: Henry Marichal, Diego Passarella, Gregory Randall
Abstract: Tree ring marking remains a key step in dendrometry and dendrochronology, but it is often performed manually, making the process time-consuming, subjective, and difficult to scale to large image datasets. We present the Tree Ring Analyzer Suite (TRAS), an open-source graphical software for automatic delineation, manual correction, and measurement of tree rings in wood cross-sectional images. TRAS integrates three complementary detection algorithms: the classical image-processing method CS-TRD and two deep-learning approaches, DeepCS-TRD and INBD. The interface allows users to refine automatic detections, remove false positives, and manually add missing rings. It also computes dendrochronological metrics such as earlywood and latewood areas, ring perimeter, equivalent ring width, and custom path-based ring-width measurements. TRAS was evaluated on 18 expertly annotated Pinus taeda L. cross-section images. DeepCS-TRD achieved the best automatic detection performance, with an F-score of 81.0% and precision of 86.4%. Automatic detection reduced the required manual correction effort to approximately 20% of ring boundaries. For one-dimensional ring-width measurements, TRAS showed excellent agreement with CooRecorder ($r > 0.99$). Common detection errors, such as jump propagation or false positives near knots, were easily corrected through the postprocessing interface. TRAS provides a flexible and reproducible solution for tree-ring analysis on Windows, macOS, and Linux. Code is available at the https://hmarichal93.github.io/tras.
Authors: Ying Shen, Tianrong Chen, Yuan Gao, Yizhe Zhang, Yuyang Wang, Miguel \'Angel Bautista, Shuangfei Zhai, Joshua M. Susskind, Jiatao Gu
Abstract: Deep generative models have advanced rapidly across text and vision, motivating unified multimodal systems that can understand, reason over, and generate interleaved text-image sequences. Most existing approaches combine autoregressive language modeling with diffusion-based image generators, inheriting a structural mismatch between causal text generation and iterative visual denoising. We observe that autoregressive normalizing flows are autoregressive Transformers--sharing the same causal mask, KV-cache mechanism, and left-to-right structure as LLMs--making them the most natural paradigm for true unified multimodal generation. We present STARFlow2, built on the Pretzel architecture that vertically interleaves a pretrained VLM stream with a TarFlow stream via residual skip connections, both operating under the same causal mask. Combined with a deep-shallow flow design and a unified FAE latent space, STARFlow2 enables cache-friendly interleaved generation where both text and visual outputs directly enter the KV-cache without re-encoding. Experiments demonstrate strong performance across image generation and multimodal understanding benchmarks, validating autoregressive flows as a viable foundation for unified multimodal modeling.
Authors: R\"uveyda Yilmaz, Yuli Wu, Johannes Stegmaier, Volkmar Schulz
Abstract: Positron Emission Tomography (PET) image reconstruction is inherently challenged by Poisson noise and physical degradation factors, which are further exacerbated in limited-angle acquisitions. While deep learning methods demonstrate promising performance, their generalization to unseen clinical data distributions remains limited without extensive retraining. We propose PET-Adapter, a test-time domain adaptation framework for generative PET reconstruction models pretrained solely on phantom data. Our method enables adaptation to clinical datasets with varying anatomies, tracers, and scanner configurations without requiring paired ground truth. PET-Adapter introduces layer-wise low-rank anatomical conditioning during adaptation and Ordered Subset Expectation Maximization-based warm-starting that initializes the generation from physics-informed reconstructions, reducing diffusion steps from 50 to 2 without compromising quality. Experiments across multiple clinical datasets demonstrate superior 3D reconstruction performance in both full-angle and limited-angle settings, highlighting the clinical feasibility and computational efficiency of the proposed approach.
Authors: Kaidi Jia, Yujie Lin, Chengyi Yang, Jiayao Ma, Jinsong Su
Abstract: Vision-language models (VLMs) raise growing concerns about privacy, copyright, and bias, motivating machine unlearning to remove sensitive knowledge. However, existing methods primarily fine-tune the language decoder, leading to superficial forgetting that fails to erase underlying visual representations and often introduces object hallucination. We propose HFRU, a reinforcement unlearning framework that operates on the vision encoder for deep semantic removal. Our two-stage approach combines alignment disruption with GRPO-based optimization using a composite reward, including an abstraction reward that encourages semantically valid substitutions and mitigates hallucinations. Experiments on object recognition and face identity tasks show that HFRU achieves over 98% forgetting and retention performance, while introducing negligible object hallucination, significantly outperforming prior methods.Our code and implementation details are available at https://github.com/XMUDeepLIT/HFRU.
Authors: Tianfei Ren, Zhipeng Yan, Yiming Zhao, Zhen Fang, Yu Zeng, Guohui Zhang, Hang Xu, Xiaoxiao Ma, Shiting Huang, Ke Xu, Wenxuan Huang, Lionel Z. Wang, Lin Chen, Zehui Chen, Jie Huang, Feng Zhao
Abstract: While text-to-image models have made strong progress in visual fidelity, faithfully realizing complex visual intents remains challenging because many requirements must be tracked across grounding, generation, and verification. We refer to these requirements as semantic commitments and formalize their lifecycle discontinuity as the Conceptual Rift, where commitments may be locally resolved or checked but fail to remain identifiable as the same operational units throughout the generation lifecycle. To address this, we propose SCOPE, a specification-guided skill orchestration framework that maintains semantic commitments in an evolving structured specification and conditionally invokes retrieval, reasoning, and repair skills around unresolved or violated commitments. To evaluate commitment-level intent realization, we introduce Gen-Arena, a human-annotated benchmark with entity- and constraint-level specifications, together with Entity-Gated Intent Pass Rate (EGIP), a strict entity-first pass criterion. SCOPE substantially outperforms all evaluated baselines on Gen-Arena, achieving 0.60 EGIP, and further achieves strong results on WISE-V (0.907) and MindBench (0.61), demonstrating the effectiveness of persistent commitment tracking for complex image generation.
Authors: Xinyan Ye, Jiankang Deng, Abbas Edalat
Abstract: Talking-head generation requires joint modeling of identity, head pose, facial expression, and mouth dynamics. Existing methods typically address only a subset of these factors, and rely on fixed-weight or heuristic fusion when multiple conditions are involved. We present MoCoTalk, a multi-conditional video diffusion framework that unifies four complementary control signals: a reference image, facial keypoints, 3DMM-rendered shading meshes, and the corresponding speech audio. To resolve destructive interference among heterogeneous conditions, we introduce an Adaptive Multi-Condition Router that computes channel-wise, timestep-aware gating over the four condition streams, allowing the fusion strategy to vary with both feature subspace and noise level. To better capture speech-related facial dynamics, we design a Mouth-Augmented Shading Mesh, a 3DMM-based representation that decouples head motion, mouth motion, expression, and lighting. This design provides a temporally consistent geometric prior and allows flexible recombination of these attributes at inference. We further introduce a lip consistency loss to tighten audio-visual alignment. Extensive experiments show that MoCoTalk achieves state-of-the-art performance on the majority of structural, motion, and perceptual metrics, while offering attribute-level controllability that single-condition methods do not provide.
Authors: Hanchao Liu, Fang-Lue Zhang, Shining Zhang, Tai-Jiang Mu, Shi-Min Hu
Abstract: Generating human motion that satisfies customized zero-shot goal functions, enabling applications such as controllable character animation and behavior synthesis for virtual agents, is a critical capability. While current approaches handle many unseen constraints, they fail on tasks with very challenging spatiotemporal restrictions, such as severe spatial obstacles or specified numbers of walking steps. To equip motion generators for these highly constrained tasks, we present a retrieval-guided method built on the training-free diffusion noise optimization framework. The key idea is to search within large motion datasets for guidance that can potentially satisfy difficult constraints. We introduce relational task parsing to group target constraints and identify the difficult ones to be handled by retrieved reference. A better initialization for diffusion noise is then obtained via a reward-guided mask that combines random noise with retrieved noise. By optimizing diffusion noise from this improved initialization, we successfully solve highly constrained generation tasks. By leveraging LLM for relational task parsing, the whole framework is further enabled to automatically reason for what to retrieve, improving the intelligence of moving agents under a training-free optimization scheme.
Authors: Ismail Aljosevic, Amir Masoud Almasi, Ana Parovic, Ashkan Shafiei
Abstract: In this paper, we propose a modular framework for 6D pose estimation based on keypoint heatmap regression. Our approach combines YOLOv10m for object detection with a ResNet18-based network that predicts 2D heatmaps from RGB images. Keypoints extracted from these heatmaps are used to estimate the 6D object pose via the PnP RANSAC algorithm. We compare different keypoint selection strategies to assess their impact on pose accuracy. Additionally, we extend the baseline by incorporating depth data using a cross-fusion architecture, which enables interaction between RGB and depth features at multiple stages. We further explore general training improvements, such as experimenting with activation functions and learning rate scheduling strategies to improve model performance. Our best RGB-only model achieved a mean ADD-based accuracy of 84.50%, while the RGB-D fusion model reached 92.41% on the LINEMOD dataset. The code is available at https://github.com/ameermasood/HeatNet.
Authors: Zhen Fang, Wenxuan Huang, Yu Zeng, Yiming Zhao, Shuang Chen, Kaituo Feng, Yunlong Lin, Lin Chen, Zehui Chen, Shaosheng Cao, Feng Zhao
Abstract: Existing Flow Matching (FM) text-to-image models suffer from two critical bottlenecks under multi-task alignment: the reward sparsity induced by scalar-valued rewards, and the gradient interference arising from jointly optimizing heterogeneous objectives, which together give rise to a 'seesaw effect' of competing metrics and pervasive reward hacking. Inspired by the success of On-Policy Distillation (OPD) in the large language model community, we propose Flow-OPD, the first unified post-training framework that integrates on-policy distillation into Flow Matching models. Flow-OPD adopts a two-stage alignment strategy: it first cultivates domain-specialized teacher models via single-reward GRPO fine-tuning, allowing each expert to reach its performance ceiling in isolation; it then establishes a robust initial policy through a Flow-based Cold-Start scheme and seamlessly consolidates heterogeneous expertise into a single student via a three-step orchestration of on-policy sampling, task-routing labeling, and dense trajectory-level supervision. We further introduce Manifold Anchor Regularization (MAR), which leverages a task-agnostic teacher to provide full-data supervision that anchors generation to a high-quality manifold, effectively mitigating the aesthetic degradation commonly observed in purely RL-driven alignment. Built upon Stable Diffusion 3.5 Medium, Flow-OPD raises the GenEval score from 63 to 92 and the OCR accuracy from 59 to 94, yielding an overall improvement of roughly 10 points over vanilla GRPO, while preserving image fidelity and human-preference alignment and exhibiting an emergent 'teacher-surpassing' effect. These results establish Flow-OPD as a scalable alignment paradigm for building generalist text-to-image models.
Authors: Jerry Jiang, Haowen Sun, Denis Gudovskiy, Yohei Nakata, Tomoyuki Okuno, Kurt Keutzer, Wenzhao Zheng
Abstract: Spatial intelligence in vision-language models (VLMs) attracts research interest with the practical demand to reason in the 3D world.Despite promising results, most existing methods follow the conventional 2D pipeline in VLMs and use pixel-aligned representations for the vision modality. However, correspondence-based models with implicit 3D scene understanding often fail to achieve spatial consistency, and representation-based models with 3D geometric priors lack efficiency in vision sequence serialization. To address this, we propose a Proxy3D method with compact yet comprehensive 3D proxy representations for the vision modality. Given only video frames as input, we employ semantic and geometric encoders to extract scene features and then perform their semantic-aware clustering to obtain a set of proxies in the 3D space. For representation alignment, we further curate the SpaceSpan dataset and apply multi-stage training to adopt the proposed 3D proxy representations with the VLM. When using shorter sequences for vision information, our method achieves competitive or state-of-the-art performance in 3D visual question answering, visual grounding and general spatial intelligence benchmarks.
Authors: Wei Yu, Yunhang Qian
Abstract: Recent event-based image reconstruction methods predominantly rely on Convolutional Neural Networks (CNNs) and Vision Transformers (ViTs) to process complementary event information. However, these architectures face fundamental limitations: CNNs often fail to capture global feature correlations, whereas ViTs incur quadratic computational complexity (e.g., $O(n^2)$), hindering their application in high-resolution scenarios. To address these bottlenecks, we introduce EmambaIR, an Efficient visual State Space Model designed for image reconstruction using spatially sparse and temporally continuous event streams. Our framework introduces two key components: the cross-modal Top-k Sparse Attention Module (TSAM) and the Gated State-Space Module (GSSM). TSAM efficiently performs pixel-level top-k sparse attention to guide cross-modal interactions, yielding rich yet sparse fusion features. Subsequently, GSSM utilizes a nonlinear gated unit to enhance the temporal representation of vanilla linear-complexity ($O(n)$) SSMs, effectively capturing global contextual dependencies without the typical computational overhead. Extensive experiments on six datasets across three diverse image reconstruction tasks - motion deblurring, deraining, and High Dynamic Range (HDR) enhancement - demonstrate that EmambaIR significantly outperforms state-of-the-art methods while offering substantial reductions in memory consumption and computational cost. The source code and data are publicly available at: https://github.com/YunhangWickert/EmambaIR
Authors: Jiatao Gu, Tianrong Chen, Ying Shen, David Berthelot, Shuangfei Zhai, Josh Susskind
Abstract: Diffusion-based models decompose sampling into many small Gaussian denoising steps -- an assumption that breaks down when generation is compressed to a few coarse transitions. Existing few-step methods address this through distillation, consistency training, or adversarial objectives, but sacrifice the likelihood framework in the process. We introduce Normalizing Trajectory Models (NTM), which models each reverse step as an expressive conditional normalizing flow with exact likelihood training. Architecturally, NTM combines shallow invertible blocks within each step with a deep parallel predictor across the trajectory, forming an end-to-end network trainable from scratch or initializable from pretrained flow-matching models. Its exact trajectory likelihood further enables self-distillation: a lightweight denoiser trained on the model's own score produces high-quality samples in four steps. On text-to-image benchmarks, NTM matches or outperforms strong image generation baselines in just four sampling steps while uniquely retaining exact likelihood over the generative trajectory.
Authors: Pengqi Lu
Abstract: Scaling Diffusion Transformers (DiTs) to hundreds of layers introduces a structural vulnerability: networks can enter a silent, mean-dominated collapse state that homogenizes token representations and suppresses centered variation. Through mechanistic auditing, we isolate the trigger event of this collapse as Mean Mode Screaming (MMS). MMS can occur even when training appears stable, with a mean-coherent backward shock on residual writers that opens deep residual branches and drives the network into a mean-dominated state. We show this behavior is driven by an exact decomposition of these gradients into mean-coherent and centered components, compounded by the structural suppression of attention-logit gradients through the null space of the Softmax Jacobian once values homogenize. To address this, we propose Mean-Variance Split (MV-Split) Residuals, which combine a separately gained centered residual update with a leaky trunk-mean replacement. On a 400-layer single-stream DiT, MV-Split prevents the divergent collapse that crashes the un-stabilized baseline; it tracks close to the baseline's pre-crash trajectory while remaining substantially better than token-isotropic gating methods such as LayerScale across the full schedule. Finally, we present a 1000-layer DiT as a scale-validation run at boundary scales, establishing that the architecture remains stably trainable at extreme depth.
Authors: Chenxi Tao, Seung-Kyum Choi
Abstract: Flow matching generates data by integrating a learned velocity field, where the number of integration steps (NFE) directly determines inference cost. We analyze which properties of the velocity field govern integration error by decomposing the velocity Jacobian into its symmetric part S (strain rate) and antisymmetric part Omega (vorticity). We prove that strain and vorticity play different roles: strain controls exponential error amplification through the logarithmic norm, while vorticity contributes only linearly to the local truncation error. We further show that the optimal transport velocity field is irrotational and has zero material derivative, implying second-order Euler accuracy; for exact displacement interpolation, the associated Lagrangian particle dynamics are integrated exactly by Euler. Motivated by this analysis, we study weighted Jacobian regularization with strain weight alpha and vorticity weight beta. Experiments on 2D synthetic data confirm the main theoretical predictions, showing up to 2.7x lower integration error at NFE=5. Preliminary CIFAR-10 experiments show consistent trends, with a lightweight fine-tuning procedure improving FID by 14 percent at NFE=10 while preserving high-NFE quality.
Authors: Lorenzo Riccardo Allegrini, Geremia Pompei
Abstract: A hierarchical ensemble pipeline is introduced to address anomaly detection in multivariate telemetry data provided by European Space Agency (ESA). The method integrates shapelet-based and statistical feature extraction, per-channel modeling, intra-channel stacking, and a final cross-channel aggregation. The pipeline is trained and validated using time-series cross-validation and two-level masking strategies to prevent information leakage. Results on the European Space Agency Anomaly Detection Benchmark (ESA-ADB) challenge demonstrate strong generalization, highlighting the effectiveness of hierarchical modeling in detecting subtle anomalies in realistic satellite telemetry.
Authors: Daniel Mensing, Jan Kapar, Jochen G. Hirsch, Matthias G\"unther, Horst Hahn, Marvin N. Wright
Abstract: We propose a multimodal latent diffusion model that jointly synthesizes volumetric magnetic resonance imaging (MRI) and tabular clinical data within a shared latent space via cross-attention. This approach enables coherent joint representation learning of MRI and tabular modalities for generative modeling. Our model utilizes a variational autoencoder to fuse the two modalities before diffusion-based synthesis, allowing modality-appropriate reconstruction with separate decoders for MRI and tabular data. We evaluated the framework on data from the German National Cohort (NAKO Gesundheitsstudie), comprising over 10,000 participants with MRI scans and clinical tabular features such as age, sex, body measurements, and ethnicity. The generated MRI volumes exhibited anatomical plausibility and body composition consistent with the synthesized tabular attributes. Quantitative evaluation using Fr\'echet distance and precision-recall metrics confirmed high-fidelity image generation. In the tabular modality, our model outperformed CTGAN across standard evaluation metrics and achieved results comparable to TVAE, demonstrating competitive performance relative to established unimodal baselines. This work is, to our knowledge, the first to demonstrate the feasibility of jointly modeling MRI and mixed-type tabular data in a single latent diffusion framework, offering a proof-of-concept for generating coherent synthetic multimodal patient data and aligning with the broader goal of developing digital twins in healthcare.
Authors: O\u{g}uzhan Fatih Kar, Roman Bachmann, Yuanzheng Gong, Anders Boesen Lindbo Larsen, Afshin Dehghan
Abstract: The web is complex, open-ended, and constantly changing, making it challenging to scale training data for visual web agents. Existing data collection attempts remain limited to offline trajectories for supervised fine-tuning or a handful of simulated environments for RL training, thus failing to capture web diversity. We propose Weblica (Web Replica), a framework for constructing reproducible and scalable web environments. Our framework leverages 1) HTTP-level caching to capture and replay stable visual states while preserving interactive behavior and 2) LLM-based environment synthesis grounded in real-world websites and core web navigation skills. Using this framework, we scale RL training to thousands of diverse environments and tasks. Our best model, Weblica-8B, outperforms open-weight baselines of similar size across multiple web navigation benchmarks while using fewer inference steps, scales favorably with additional test-time compute, and is competitive with API models.
Authors: Hashim Aziz, Mehedi Hasan Raju, Oleg V. Komogortsev
Abstract: Eye movement biometrics (EMB) use subject-specific gaze dynamics for user authentication and identification. Recent deep learning-based EMB systems achieve strong performance by modeling temporal eye movement behavior. However, these systems typically overlook continuous gaze offset, despite prior evidence that it contains user-discriminative information. This work examines whether continuous gaze offset can improve biometric performance when combined with existing biometric features. We evaluate linear and nonlinear fusion methods on two publicly available datasets, collected via the lab-grade eye tracker and virtual reality headset across multiple tasks and observation durations. Results indicate that fusion offers performance benefits on both datasets, particularly when using nonlinear fusion. Additionally, fusing biometric information across multiple tasks further improves authentication performance. These findings support the hypothesis that continuous gaze offset may serve as useful auxiliary information under conditions of degraded or noisy eye tracking.
Authors: Isaac Galatzer-Levy, Daniel McDuff, Xin Liu, Jed McGiffin
Abstract: The pursuit of artificial general intelligence necessitates robust methods for evaluating the cognitive capabilities of models beyond narrow task performance. Here, we introduce a psychometric framework to assess the cognitive profiles of generative AI, comparing them to human norms and tracking their evolution across generations. Initial evaluation of leading multimodal models using tasks adapted from the Wechsler Adult Intelligence Scale revealed a profoundly uneven cognitive architecture: near-ceiling performance in verbal comprehension and working memory (>$98^{\text{th}}$ percentile) contrasted with near-floor performance in perceptual reasoning (<$1^{\text{st}}$ percentile). To track developmental trajectories beyond human-normed limits, we developed the Artificial Intelligence Quotient (AIQ) Benchmark and applied it to six generations and two model families, revealing significant but asymmetric performance gains. Notably, we uncovered a sharp dissociation between modalities; abstract quantitative reasoning matured far more rapidly when presented linguistically compared to a visually analogous format, indicating an architectural bias towards language-based symbolic manipulation. While abstract visual reasoning improved, visual-perceptual organization remained largely stagnant. Collectively, these findings demonstrate that the cognitive abilities of generative models are evolving unevenly, suggesting that scaling and optimization approaches to AGI development alone may be insufficient to overcome fundamental architectural limitations in achieving balanced, human-like general intelligence.
Authors: Aditya Ranganath, Mukesh Singhal
Abstract: We survey continuous-time generative modeling methods based on transporting a simple reference distribution to a data distribution via stochastic or deterministic dynamics. We present a unified framework in which diffusion models, score-based generative models, and flow matching are instances of learning a time-dependent vector field that induces a family of marginals $(\rho_t)_{t \in [0,1]}$ governed by continuity and Fokker-Planck equations. Such a unified theory is timely because these methods are converging methodologically, yet fragmented notation and competing derivations continue to obscure their shared structure and the practical tradeoffs governing sampling, stability, and computation. Within this framework, we (i) derive reverse-time sampling for diffusion and score-based models as controlled stochastic dynamics, (ii) show that the probability flow ODE yields identical marginals and connects diffusion to likelihood-based normalizing flows, and (iii) interpret flow matching as direct regression of the velocity field under a chosen interpolation, clarifying when it coincides with or differs from score-based training. We compare objectives, sampling schemes, and discretization errors under unified notation, discuss connections to Schrodinger bridges and entropic optimal transport, and summarize theoretical guarantees and open problems on approximation, stability, and scalability.
Authors: Mukul Lokhande, Ratko Pilipovic, Omkar Kokane, Adam Teman, Santosh Kumar Vishvakarma
Abstract: Advanced driver-assistance systems (ADAS) require neural compute engines that deliver low-latency inference under strict power and area constraints. Posit arithmetic is attractive for such accelerators because it provides high numerical fidelity at low precision, but its variable-length regime encoding increases encode/decode cost and exposes the datapath to large regime-field fault effects. This paper presents EULER-ADAS, a SIMD-enabled logarithmic bounded-Posit neural compute engine for energyefficient and reliability-aware ADAS acceleration. The proposed datapath combines bounded-regime Posit representation, stageadaptive logarithmic mantissa multiplication with bit truncation, and a SIMD-shared quire accumulation path supporting Posit- (8,0), Posit-(16,1), and Posit-(32,2) execution. The unified architecture enables 4xPosit-8, 2xPosit-16, or 1xPosit-32 operation without duplicating precision-specific hardware. FPGA implementation shows that the proposed configurations reduce LUT count by up to 41.4%, delay by up to 76.1%, and power by up to 71.9% relative to exact Posit neural compute engines, while achieving up to 10x lower energy-delay product than radix-4 Booth-based Posit multipliers. In 28-nm CMOS, the bounded variants occupy 0.013-0.016 mm2 , consume 19.8-22.1 mW, and operate at up to 1.84 GHz. Application-level evaluation across image-classification, ADAS, and edge-inference workloads shows that the evaluated Posit-16 and Posit-32 configurations remain within about 1.5 percentage points of FP32 accuracy. A TinyYOLOv3 prototype on Pynq-Z2 achieves 78 ms latency at 0.29 W and 22.6 mJ/frame, demonstrating the suitability of EULERADAS for low-power real-time ADAS inference.
Authors: Daniil Lisus, Cedric Le Gentil, Timothy D. Barfoot
Abstract: This paper introduces Dr-BA, a first-of-its-kind radar bundle adjustment (BA) framework that operates directly on 2D spinning radar intensity images. Unlike camera or lidar sensors, radar is largely unaffected by precipitation, making it a critical modality for autonomous systems that require all-weather robustness. Existing state estimation approaches using spinning radar typically extract sparse point clouds from range-azimuth-intensity measurements and apply point cloud alignment techniques to estimate vehicle motion, scene structure, or to localize within an existing map. In contrast, Dr-BA uses the full radar returns from multiple scans to jointly estimate dense maps and sensor poses. By formulating the problem as a separable optimization, we derive an efficient and general solution that decouples pose estimation from mapping. In addition to solving the BA problem, this formulation naturally extends to direct radar-only localization (DRL) within a previously built map. Dr-BA achieves state-of-the-art radar-based BA and cross-session localization performance, demonstrated on more than 200 km of on-road data across five distinct routes. Our implementation is publicly available at https://github.com/utiasASRL/dr_ba.
Authors: Zijun Cui, Xiulong Liu, Hao Fang, Mingwei Xu, Jiageng Liu, Zexin Xu, Weiguo Pian, Shijian Deng, Feiyu Du, Chenming Ge, Yapeng Tian
Abstract: Joint audio-video generation models are rapidly approaching professional production quality, raising a central question: do they understand audio-visual physics, or merely generate plausible sounds and frames that violate real-world consistency? We introduce AV-Phys Bench, a benchmark for evaluating physical commonsense in joint audio-video generation. AV-Phys Bench tests models across three scene categories: Steady State, Event Transition, and Environment Transition. It covers physics-grounded subcategories drawn from real-world scenes, plus Anti-AV-Physics prompts that deliberately request physically inconsistent audio-video behavior. Each generation is evaluated along five dimensions: visual semantic adherence, audio semantic adherence, visual physical commonsense, audio physical commonsense, and cross-modal physical commonsense. Across three proprietary and four open-source models, we find that Seedance 2.0 performs best overall, but all models remain far from robust physical understanding. Performance drops sharply on event-driven and environment-driven transitions, and even strong proprietary systems collapse on Anti-AV-Physics prompts. We further introduce AV-Phys Agent, a ReAct-style evaluator that combines a multimodal language model with deterministic acoustic measurement tools, producing rankings that closely align with human ratings. Our results identify cross-modal physical consistency and transition-driven scene dynamics as key open challenges for joint audio-video generation.
Authors: Quanliang Liu, Jungtaek Kim, Kangwook Lee, Hyunseok Oh
Abstract: Vision-language models (VLMs) have shown strong potential for scientific image understanding, but general-purpose models often lack the domain-specific visual knowledge required for reliable materials characterization. In this work, we fine-tuned an open-source VLM (Qwen3-VL-32B-Instruct) for fracture-surface image analysis using a curated dataset of 13,168 open-source, literature-mined fracture-surface images. Morphology annotations were generated by GPT-5.2-Reasoning (high) from both the images and relevant excerpts of their source papers, and the dataset was further enriched with targeted manual collection and rotation-based augmentation. The resulting specialist model outperforms flagship proprietary multimodal models on a benchmark of 100 manually annotated images. It achieves a precision of 0.92, compared to 0.35 for the base Qwen3-VL-32B-Instruct, 0.58 for GPT-5.5-Reasoning (high), and 0.78 for Gemini 3.1 Pro-Reasoning (high). Dataset ablations show that manual collection of rare-feature images and augmentation via image rotation are both beneficial to improve recognition of less common fracture morphology features. We further discuss integrated use of the fine-tuned model with proprietary models to combine fracture-specific visual accuracy with broader multimodal reasoning for autonomous fractography. Although focused on fracture-surface images, this work demonstrates how VLMs can be adapted through targeted collection and fine-tuning on novel feature images to recognize those features and support downstream decision-making in autonomous microscopy workflows.
Authors: Junchuan Zhao, Qifan Liang, Ye Wang
Abstract: Co-speech gesture generation aims to synthesize realistic body movements that are semantically coherent with speech and faithful to a user-specified gestural style. Existing VQ-VAE based co-speech gesture generation methods improve generation quality but fail to encode semantic structure into the motion representation or explicitly disentangle content from style, limiting both semantic coherence and personalization fidelity. We present PersonaGest, a two-stage framework addressing both limitations. In the first stage, a semantic-guided RVQ-VAE disentangles motion content and gestural style within the residual quantization structure, where a Semantic-Aware Motion Codebook (SMoC) organizes the content codebook by gesture semantics and contrastive learning further enforces content-style separation. In the second stage, a Masked Generative Transformer generates content tokens via a semantic-aware re-masking strategy, followed by a cascade of Style Residual Transformers conditioned on a reference motion prompt for style control. Extensive experiments demonstrate state-of-the-art performance on objective metrics and perceptual user studies, with strong style consistency to the reference prompt. Our project page with demo videos is available at https://danny-nus.github.io/PersonaGest/
Authors: Wenyuan Li, Guang Li, Keisuke Maeda, Takahiro Ogawa, Miki Haseyama
Abstract: A latent world model may achieve accurate short-horizon prediction while still inducing a latent space that is poorly aligned with planning. A key issue is spatiotemporal mismatch: these models are often trained with local predictive supervision, but deployed for long-horizon goal-directed search in latent spaces where Euclidean distance may not reflect what is reachable within a finite action budget. We present the Reachability-Correction auxiliary objective (RC-aux), a lightweight correction for this mismatch in reconstruction-free latent world models. RC-aux keeps the world-model backbone unchanged and adds planning-aligned supervision along two axes. Along the time axis, multi-horizon open-loop prediction trains the model beyond one-step consistency. Along the space axis, budget-conditioned reachability supervision, together with temporal hard negatives, encourages the latent space to distinguish states that are eventually reachable from those reachable within the current planning horizon. At test time, the learned reachability signal can also be used by a reachability-aware planner to favor trajectories that are both goal-directed and attainable under the available budget. We instantiate RC-aux on LeWorldModel and evaluate it under both continuation-training and matched-from-scratch settings. Across goal-conditioned pixel-control tasks and a LIBERO-Goal extension, RC-aux improves LeWM-style planning with modest additional cost. These results suggest that planning with latent world models depends not only on predictive accuracy, but also on whether the learned representation encodes the temporal and geometric structure required by downstream search. The code is available at https://github.com/Guang000/RC-aux.
Authors: Jingyi Liu, Cheng Yuan, Lijun He, Jun Zhang, Jiawei Shao
Abstract: The expanding application of smart sensing has created a growing demand for the accurate understanding of human action at the network edge. Traditional approaches require massive video data to be transmitted from resource-constrained edge devices to powerful cloud servers, incurring prohibitive uplink bandwidth consumption and unacceptable latency while raising privacy concerns. To overcome these bottlenecks, we propose a task-oriented communication framework for human action understanding (TOAU) through edge-cloud collaboration. Our framework utilizes a monocular pose estimator to extract continuous joint coordinates from raw videos, followed by a vector quantized variational autoencoder (VQ-VAE) to convert these coordinates into discrete motion tokens. Consequently, only a compact sequence of codebook indices is transmitted over the network, consuming as few as 9 bits per frame and avoiding privacy leakages. At the cloud server, a lightweight projector aligns these motion tokens with the embedding space of a large vision-language model (VLM) to facilitate complex action understanding, which is trained with an efficient instruction tuning paradigm. Comprehensive evaluations on three benchmarks demonstrate that our TOAU system reduces the transmission payload to approximately 1\% and the system latency to around 20\% compared to video codec-based solutions, while delivering comparable action understanding accuracy.
Authors: Kali Hamilton, Christoffer Heckman
Abstract: Cameras and LiDAR degrade in rain, fog, and snow, while millimeter-wave radar remains largely unaffected. We align a radar encoder to frozen SigLIP vision embeddings and decode structured scene captions through a frozen vision-language model (VLM) with approximately 7M trainable parameters. On K-RADAR with held-out fog, light snow, and heavy snow sequences, all radar configurations outperform a camera baseline that collapses to over 90% hallucination. We identify a token-norm mismatch as the dominant failure mode when bridging radar to a frozen VLM and show that projector-output LayerNorm resolves it. Analysis of encoder complexity, caption format, and pooling strategy reveals tradeoffs that inform future radar-VLM pipeline design.
Authors: Hao Liu, Yuxuan Lin, Jingfeng Guo, Ruihang Chu, Junjie Wang, Ruotong Li, Yujiu Yang
Abstract: Editing a 3D asset locally, modifying a target region while preserving the rest, is a fundamental requirement of native 3D editing. Existing methods enforce locality through mechanisms external to the generator, such as manual 3D masks, post-hoc voxel merging, or 2D multi-view lifting. None of them intervene where the corruption actually originates: inside the ODE sampler. For a rectified-flow generator to achieve faithful local editing, its velocity field should be strong over the target editing region while vanishing on preserved content. Yet a single velocity field can hardly satisfy both requirements simultaneously, leading to three problems: (i) identity leakage that keeps the edit signal non-zero on preserved regions; (ii) no dedicated edit-amplification channel, so strengthening the edit inevitably perturbs identity; and (iii) an identity drag at the geometry and material stages, where a global condition pulls every token toward the target. We propose VS3D (Velocity-Space 3D Asset editing}), an inversion-free, training-free, and mask-free framework that addresses each problem with a targeted intervention inside the sampler. VS3D integrates three complementary modules, each corresponding to a specific stage of the editing pipeline. Reconstruction-Anchored Source Injection (RASI) absorbs identity leakage by turning the unconditional embedding into a per-step, asset-specific anchor calibrated through source reconstruction. Partial-Mean Guidance (PMG) amplifies the edit signal by contrasting high- and low-quality subsample estimates of the velocity difference, active only where a consistent edit exists. Twin-Agreement Residual injection (TAR) lets the sampler decide token by token what to preserve at the geometry and material stages.
Authors: Fengqiang Wan, Yipeng Lin, Kan Lv, Yang Yang
Abstract: Pre-trained models with parameter-efficient fine-tuning (PEFT) have demonstrated promising potential for class-incremental learning (CIL), yet catastrophic forgetting still persists when adapting models to new tasks. In this paper, we present a novel perspective on catastrophic forgetting through the analysis of inter-layer relation drift, i.e., the progressive disruption of relationships among layer-wise representations during the learning of new tasks. We theoretically show that the increase of such drift reduces the classification margins of previously learned tasks, thereby degrading overall model performance. To address this issue, we propose \underline{S}elf-\underline{R}ectifying inter-layer \underline{R}elation Low-Rank Adaptation~(SR$^2$-LoRA), a simple yet effective method that mitigates catastrophic forgetting by constraining inter-layer relation drift. Specifically, SR$^2$-LoRA constructs the relation matrices induced by the previous and current models on current-task samples, and aligns the corresponding singular values. We further theoretically show that this alignment exhibits greater robustness to estimation perturbations than direct entry-wise alignment. Extensive experiments on standard CIL benchmarks demonstrate that SR$^2$-LoRA effectively mitigates catastrophic forgetting, with its advantages becoming more pronounced as the number of tasks increases. Code is available in the \href{https://github.com/FqWan24/SR-2-LoRA}{repository}.
Authors: Bo-Kai Ruan, Teng-Fang Hsiao, Ling Lo, Hong-Han Shuai
Abstract: World Action Models (WAMs) enable decision-making through imagined rollouts by predicting future observations and actions. However, the reliability of these imagined futures remains under-examined: is a generated future merely visually plausible, or is it dynamically compatible with the action sequence it claims to model? In this work, we identify action-state consistency, the alignment between predicted actions and induced state transitions, as a missing reliability axis for WAMs. Through a systematic study across representative joint-prediction and inverse-dynamics models, we find that action-state consistency systematically separates successful and failed rollouts across many tasks and follows similar success-failure trends as learned value estimates. These results suggest that consistency captures decision-relevant structure beyond visual realism. We further identify background collapse as an important boundary condition, where low-dynamics failed trajectories can become deceptively consistent because static futures are easier to predict. Building on these findings, we introduce a value-free consensus strategy for test-time selection, which ranks candidate rollouts by agreement among predicted futures. This strategy improves success rates on RoboCasa and RoboTwin 2.0 without additional training or reward modeling. Taken together, our findings establish action-state consistency as both a diagnostic tool for evaluating WAM reliability and a practical signal for value-free planning.
Authors: George Rapakoulias, Peter Garud, Lingjiong Zhu, Panagiotis Tsiotras
Abstract: Diffusion models achieve strong generation quality, diversity, and distribution coverage, but their performance often comes with expensive inference. In this work, we propose Stochastic Transition-Map Distillation (STMD), a teacher-free framework for accelerating diffusion model inference while preserving probabilistic sample generation. In contrast to score-based diffusion models, whose denoising parametrization models the mean of the posterior distribution, STMD distills the full transition map associated with the sampling stochastic differential equation (SDE). We parameterize these SDE transitions with a conditional Mean Flow model, yielding a one- or few-step stochastic sampler that retains the transition structure of the underlying diffusion process. This perspective is especially useful for downstream tasks that require stochastic inference, such as diffusion posterior sampling, inverse problems, and energy-based fine-tuning. Compared to recent distillation methods, STMD requires no pretrained teacher, bi-level optimization, or trajectory simulation and caching, enabling efficient and scalable training. We derive convergence bounds for our method in the Wasserstein distance, providing a strong theoretical foundation for our approach, and validate STMD on various image generation examples on the MNIST, CIFAR-10, and CelebA datasets.
Authors: Hugo Vigna, Samuel Bontemps
Abstract: The Hessian spectrum of trained deep networks exhibits a characteristic structure: a continuous bulk of near-zero eigenvalues and a small number of large outlier eigenvalues (spikes), confirming the relevance of Random Matrix Theory in deep learning. The spike count matches the number of classes minus one. While prior work has described this structure, no method has exploited it operationally to improve classification performance. We propose Spectral Surgery, a post-hoc optimization method that directly perturbs model weights along spike eigenvectors to rebalance per-class accuracy without retraining. We introduce (i) a spike-class sensitivity matrix that quantifies the directional derivative of each class's accuracy along each spike eigenvector, (ii) a constrained optimization of perturbation coefficients that targets weak classes while preserving strong ones, and (iii) an adaptive amplitude control that raises or lowers the perturbation budget based on iteration-level improvement signals. We obtain encouraging results on CIFAR-10 and ISIC-2019 on both balanced accuracy and standard deviation.
Authors: Xudong Lv, Yuxiang Sun, Shuo Wang, Nanxing Chen, Jun Guan, Jingtian Hu
Abstract: Optical neural networks are emerging as powerful machine learning and information processing tools because of their potential advantages in speed and energy efficiency. The training methods of these physical models, however, remain underexplored compared to their digital counterparts and are leading to suboptimal performance. This paper reports a pre-training-driven approach that leads to snapshot image denoising with substantially improved quality. We demonstrated effective free-space optical denoising by a diffractive network optimized by a two-step process including (1) pre-training using a massive dataset of 3.45 million diverse but simple images and (2) fine-tuning with the corresponding task-specific datasets. Compared to conventional Fourier-domain filtering and directly trained diffractive networks, such a transfer learning process exhibited prominent advantages for denoising images degraded by severe noise, peak signal-to-noise ratio (PSNR) below 8 dB, while preserving fine image features and improving the PSNR to above 18 dB. Importantly, the same pre-trained optical network could be consistently fine-tuned to process degraded images from highly diverse styles ranging from handwritten digits (MNIST) and chest X-rays (ChestMNIST) to CIFAR-10 images and human faces (CelebA). We further demonstrated the critical role of our optical denoisers in vision-based applications, including face detection, plate recognition, and localization of UAVs in noisy conditions.
Authors: Xiaomin Yu, Yijiang Li, Yuhui Zhang, Hanzhen Zhao, Yue Yang, Hao Tang, Yue Song, Xiaobin Hu, Chengwei Qin, Shuicheng Yan, Hui Xiong
Abstract: Training multimodal large language models has long been limited by the scarcity of high-quality paired multimodal data. Recent studies show that the shared representation space of pretrained multimodal contrastive models can serve as a bridge, enabling models to perform multimodal training with unimodal data. However, the key premise of this paradigm remains insufficiently understood: can representations from different modalities be reliably interchanged? The core obstacle lies in the persistent Modality Gap in the shared space. In this work, we revisit the geometric nature of the modality gap. We find that modality representations already share compatible dominant semantic geometry. What truly hinders modality interchangeability is not a simple global shift, but an anisotropic residual structure concentrated along a small number of dominant directions. Based on this finding, we further propose the principle of anisotropic modality gap alignment: effective modality alignment should align with the target-modality distribution while preserving the semantic structure of the source modality. Guided by this principle, we propose an anisotropic geometric correction framework, AnisoAlign, for unpaired modality alignment. This framework leverages the internal geometric prior of the target modality and performs bounded correction on source-modality representations, thereby constructing substitute representations in the target modality. Experiments confirm its benefits in both geometric diagnostics and text-only MLLM training. Overall, this work recasts the modality gap from an empirical observation into a correctable, structured geometric phenomenon and provides a new representation alignment perspective for training multimodal models with unimodal data.
Authors: Ozgu Goksu, Nicolas Pugeault
Abstract: Federated Learning (FL) enables decentralised model training across distributed clients without requiring data centralisation. However, the generalisation performance of the global model is usually degraded by data heterogeneity across clients, particularly under limited data availability and class imbalance. To address this challenge, we propose FedQuad, a novel method that explicitly enforces minimising intra-class representations while enabling inter-class splits across clients. By jointly minimising distances between positive pairs and maximising distances between negative pairs, the proposed approach mitigates representation misalignment introduced during model aggregation. We evaluate our method on CIFAR-10, CIFAR-100, and Tiny-ImageNet under diverse non-IID settings and varying numbers of clients, demonstrating consistent improvements over existing baselines. Additionally, we provide a comprehensive analysis of metric learning-based approaches in both centralised and federated environments, highlighting their effectiveness in alleviating representation collapse under heterogeneous data distributions.
Authors: Alessio Spagnoletti, Tim Y. J. Wang, Marcelo Pereyra, O. Deniz Akyildiz
Abstract: Vision-Language Latent Diffusion Models (LDMs) (Rombach et al., 2022) provide powerful generative priors for inverse problems. However, existing LDM-based inverse solvers typically require a large number of neural function evaluations (NFEs) and backpropagation through large pretrained components, leading to substantial computational costs and, in some cases, degraded reconstruction quality. We propose a unified Euclidean-Wasserstein-2 gradient-flow framework that jointly performs posterior sampling and prompt optimization in the latent space through a single flow that aligns the prior and posterior with the observed data. Combined with few-step latent text-to-image models, this formulation enables low-NFE inference without backpropagation through autoencoders. Experiments across several canonical imaging inverse problems show that our method achieves state-of-the-art performance with significantly reduced computational cost.
Authors: Aristotelis Ballas, Christos Diou
Abstract: Sharpness-aware and gradient-alignment methods have been shown to improve generalization, however each family of methods targets a single geometric property of the loss landscape, while ignoring the other. In this paper, we show that this omission is structurally unavoidable and that both flatness and gradient alignment should be considered in multi-distribution learning settings. Specifically, we derive an excess-risk decomposition that yields two additive leading-order terms: (i) an alignment term, controlled by the trace of $\bar{H}^{-1}\Sigma_g$ and (ii) a curvature term, controlled by $\bar{H}$, where $\bar{H}$ is the average Hessian and $\Sigma_g$ is the covariance of the gradient across distributions. Notably, $\bar{H}$ appears inverted in one and non-inverted in the other. We further show, via a counterexample, that neither quantity bounds the other in general, so no algorithm targeting only one term can guarantee low excess risk. Motivated by this decomposition, we propose SAGE (Spectral-Aware Gradient-Aligned Exploration) that targets both terms. The curvature component replaces SAM's gradient-scaled perturbation with the polar factor of each layer's gradient matrix, computed via Newton-Schulz iteration, so that the ascent step probes all directions with similar magnitude. On the other hand, the alignment component injects isotropic noise at the descent step, the magnitude of which scales with cross-distribution gradient disagreement. Experiments on five domain-generalization and two multi-task learning benchmarks show that the proposed method establishes a new state-of-the-art on DomainBed and acts as a general-purpose improvement to base MTL solvers, remaining competitive with, or even surpassing, state-of-the-art methods.
Authors: Giacomo Spigler
Abstract: Active vision -- where a policy controls its own gaze during manipulation -- has emerged as a key capability for imitation learning, with multiple independent systems demonstrating its benefits in the past year. Yet there is no shared benchmark to compare approaches or quantify what active vision contributes, on which task types, and under what conditions. We introduce TAVIS, evaluation infrastructure for active-vision imitation learning, with two complementary task suites -- TAVIS-Head (5 tasks, global search via pan/tilt necks) and TAVIS-Hands (3 tasks, local occlusion via wrist cameras) -- on two humanoid torso embodiments (GR1T2, Reachy2), built on IsaacLab. TAVIS provides three evaluation primitives: a paired headcam-vs-fixedcam protocol on identical demonstrations; GALT (Gaze-Action Lead Time), a novel metric grounded in cognitive science and HRI that quantifies anticipatory gaze in learned policies; and procedural ID/OOD splits. Baseline experiments with Diffusion Policy and $\pi_0$ reveal that (i) active-vision generally helps, but benefits are task-conditional rather than uniform; (ii) multi-task policies degrade sharply under controlled distribution shifts on both suites; and (iii) imitation alone yields anticipatory gaze, with median lead times comparable to the human teleoperator reference. Code, evaluation scripts, demonstrations (LeRobot v3.0; ~2200 episodes) and trained baselines are released at https://github.com/spiglerg/tavis and https://huggingface.co/tavis-benchmark.
URLs: https://github.com/spiglerg/tavis, https://huggingface.co/tavis-benchmark.
Authors: Jan Verh\"ulsdonk, Thomas Grandits, Francisco Sahli Costabal, Thomas Beiert, Simone Pezzuto, Alexander Effland
Abstract: Atlas-based approaches allow high-quality, patient-specific shape reconstructions of cardiac anatomy from sparse and/or noisy data such as point clouds. However, these methods are mainly prior-driven, so the impact of uncertainty can be large, limiting their clinical reliability. We propose a probabilistic framework for uncertainty-aware cardiac shape reconstruction that combines Deep Signed Distance Functions (DeepSDFs) with Markov Chain Monte Carlo (MCMC) sampling. Cardiac geometries are modeled implicitly as zero-level sets of a neural network conditioned on learned latent codes, enabling multi-surface reconstruction of the left and right ventricles. By interpreting the reconstruction loss as a log-likelihood, we perform Bayesian inference in the latent space to obtain both maximum a posteriori (MAP) and posterior-sampled reconstructions. Experiments on a public cardiac dataset show that our approach produces accurate reconstructions and well-calibrated uncertainty estimates.
Authors: Daniel Dauner, Valentin Charraut, Bastian Berle, Tianyu Li, Long Nguyen, Jiabao Wang, Changhui Jing, Maximilian Igl, Holger Caesar, Boris Ivanovic, Yiyi Liao, Andreas Geiger, Kashyap Chitta
Abstract: The pursuit of autonomous driving has produced one of the richest sensor data collections in all of robotics. However, its scale and diversity remain largely untapped. Each dataset adopts different 2D and 3D modalities, such as cameras, lidar, ego states, annotations, traffic lights, and HD maps, with different rates and synchronization schemes. They come in fragmented formats requiring complex dependencies that cannot natively coexist in the same development environment. Further, major inconsistencies in annotation conventions prevent training or measuring generalization across multiple datasets. We present 123D, an open-source framework that unifies such multi-modal driving data through a single API. To handle synchronization, we store each modality as an independent timestamped event stream with no prescribed rate, enabling synchronous or asynchronous access across arbitrary datasets. Using 123D, we consolidate eight real-world driving datasets spanning 3,300 hours and 90,000 kilometers, together with a synthetic dataset with configurable collection scripts, and provide tools for data analysis and visualization. We conduct a systematic study comparing annotation statistics and assessing each dataset's pose and calibration accuracy. Further, we showcase two applications 123D enables: cross-dataset 3D object detection transfer and reinforcement learning for planning, and offer recommendations for future directions. Code and documentation are available at https://github.com/kesai-labs/py123d.
Authors: Jingyun Wang, Guoliang Kang
Abstract: Recent works utilize CLIP to perform the challenging unsupervised semantic segmentation task where only images without annotations are available. However, we observe that when adopting CLIP to such a pixel-level understanding task, unexpected bias (including class-preference bias and space-preference bias) occurs. Previous works don't explicitly model the bias, which largely constrains the segmentation performance. In this paper, we propose to explicitly model and rectify the bias existing in CLIP to facilitate the unsupervised semantic segmentation task. Specifically, we design a learnable "Reference" prompt to encode class-preference bias and a projection of the positional embedding in the vision transformer to encode space-preference bias respectively. To avoid interference, two kinds of biases are firstly independently encoded into different features, i.e., the Reference feature and the positional feature. Via a matrix multiplication between the Reference feature and the positional feature, a bias logit map is generated to explicitly represent two kinds of biases. Then we rectify the logits of CLIP via a simple element-wise subtraction. To make the rectified results smoother and more contextual, we design a mask decoder which takes the feature of CLIP and the rectified logits as input and outputs a rectified segmentation mask with the help of Gumbel-Softmax operation. A contrastive loss based on the masked visual features and the text features of different classes is imposed, which makes the bias modeling and rectification process meaningful and effective. Extensive experiments on various benchmarks including PASCAL VOC, PASCAL Context, ADE20K, Cityscapes, and COCO Stuff demonstrate that our method performs favorably against previous state-of-the-arts. The implementation is available at: https://github.com/dogehhh/ReCLIP.
Authors: Haojie Zhang, Zhihao Liang, Ruibo Fu, Bingyan Liu, Zhengqi Wen, Xuefei Liu, Jianhua Tao, Yaling Liang
Abstract: Long-duration talking video synthesis faces enduring challenges in achieving high video quality, portrait consistency, temporal coherence, and computational efficiency. As video length increases, issues such as visual degradation, portrait drift, temporal artifacts, and error accumulation become increasingly problematic, severely affecting the realism and reliability of the results. To address these challenges, we present LetsTalk, a diffusion transformer framework equipped with multimodal guidance and a novel memory bank mechanism, explicitly maintaining contextual continuity and enabling robust, high-quality, and efficient generation of long-duration talking videos. In particular, LetsTalk introduces a noise-regularized memory bank to alleviate error accumulation and sampling artifacts during extended video generation. To further improve efficiency and spatiotemporal modeling, LetsTalk employs a deep compression autoencoder and a spatiotemporal-aware transformer with linear attention for effective multimodal fusion. We systematically analyze three fusion schemes and show that combining deep (Symbiotic Fusion) for portrait features and shallow (Direct Fusion) for audio achieves superior visual realism and precise speech-driven motion, while preserving diversity of movements. Extensive experiments demonstrate that LetsTalk establishes new state-of-the-art in generation quality, producing temporally coherent and realistic talking videos with enhanced diversity and liveliness, and maintains remarkable efficiency with 8x fewer parameters than previous approaches.
Authors: Aneeq Zia, Max Berniker, Rogerio Nespolo, Xiaorui Zhang, Conor Perreault, Ziheng Wang, Benjamin Mueller, Ryan Schmidt, Kiran Bhattacharyya, Xi Liu, Anthony Jarc
Abstract: Owing to recent advances in machine learning and the ability to harvest large amounts of data during robotic-assisted surgeries, surgical data science is ripe for foundational work. We present a large dataset of surgical videos and their accompanying labels for this purpose. We describe how the data was collected and some of its unique attributes. Multiple example problems are outlined. Although the dataset was curated for a particular set of scientific challenges (in an accompanying paper), it is general enough to be used for a broad range machine learning questions. Our hope is that this dataset exposes the larger machine learning community to the challenging problems within surgical data science, and becomes a touch-stone for future research. The videos are available at https://storage.googleapis.com/isi-surgvu/surgvu24_videos_only.zip, the labels at https://storage.googleapis.com/isi-surgvu/surgvu24_labels_updated_v2.zip, a validation set for tool detection problem at https://storage.googleapis.com/isi-surgvu/cat1_test_set_public.zip, and a sample set of question & answer pairs dataset for surgical visual question answering at https://storage.googleapis.com/isi-surgvu/SURGVU25_cat_2_sample_set_public.zip.
URLs: https://storage.googleapis.com/isi-surgvu/surgvu24_videos_only.zip,, https://storage.googleapis.com/isi-surgvu/surgvu24_labels_updated_v2.zip,, https://storage.googleapis.com/isi-surgvu/cat1_test_set_public.zip,, https://storage.googleapis.com/isi-surgvu/SURGVU25_cat_2_sample_set_public.zip.
Authors: Ruofan Wang, Xingjun Ma
Abstract: Large Vision-Language Models (VLMs) are increasingly deployed in open-ended environments, where ensuring reliable safety under multimodal inputs is critical. However, existing evaluations remain largely instruction-centric, focusing on explicit malicious queries while overlooking a more realistic and underexplored risk: whether safety alignment remains robust under harmful contextual exposure. This limitation is particularly important for multimodal systems, where visual inputs can substantially steer model behavior and render text-only auditing insufficient. In this work, we study multimodal safety auditing under harmful contextual exposure, asking whether VLMs can maintain safe behavior when partial toxic text is paired with visual context. To enable systematic auditing, we propose RedDiffuser (RedDiff), a reinforcement-based framework that leverages diffusion models to generate semantically coherent visual inputs for black-box safety testing. By combining greedy prompt search with reinforcement optimization, RedDiffuser uncovers high-risk multimodal inputs that expose latent safety failures. Extensive experiments on both open-source and commercial VLMs show that such context-conditioned failures are widespread. On LLaVA, RedDiffuser increases unsafe response rates by up to 10.69% on the original set and 8.91% on a hold-out set, with strong transferability to Gemini and LLaMA-Vision. These vulnerabilities persist even under external safety guardrails, suggesting that current system-level safety mechanisms remain insufficient for realistic multimodal risks. Our findings reveal a critical blind spot in existing safety evaluations and establish context-aware multimodal auditing as an essential paradigm for diagnosing hidden vulnerabilities in modern VLM systems.
Authors: Marta Hasny, Maxime Di Folco, Keno Bressem, Julia Schnabel
Abstract: Contrastive learning methods in computer vision typically rely on augmented views of the same image or multimodal pretraining strategies that align paired modalities. However, these approaches often overlook semantic relationships between distinct instances, leading to false negatives when semantically similar samples are treated as negatives. This limitation is especially critical in medical imaging domains such as cardiology, where demographic and clinical attributes play a critical role in assessing disease risk and patient outcomes. We introduce a tabular-guided contrastive learning framework that leverages clinically relevant tabular data to identify patient-level similarities and construct more meaningful pairs, enabling semantically aligned representation learning without requiring joint embeddings across modalities. Additionally, we adapt the k-NN algorithm for zero-shot prediction to overcome the lack of zero-shot capability in unimodal representations. We demonstrate the strength of our methods using a large cohort of short-axis cardiac MR images and clinical attributes, where tabular data helps to more effectively distinguish between patient subgroups. Evaluation on downstream tasks, including fine-tuning, linear probing, and zero-shot prediction of cardiovascular artery diseases and cardiac phenotypes, shows that incorporating tabular data guidance yields stronger visual representations than conventional methods that rely solely on image augmentation or combined image-tabular embeddings. Further, we show that our method can generalize to natural images by evaluating it on a car advertisement dataset. Code is available at https://github.com/marteczkah/tables_guide_vision.
Authors: Gaspard Goupy, Pierre Tirilly, Ioan Marius Bilasco
Abstract: Direct training of Spiking Neural Networks (SNNs) on neuromorphic hardware can greatly reduce energy costs compared to GPU-based training. However, implementing Backpropagation (BP) on such hardware is challenging because forward and backward passes are typically performed by separate networks with distinct weights. To compute correct gradients, forward and feedback weights must remain symmetric during training, necessitating weight transport between the two networks. This symmetry requirement imposes hardware overhead and increases energy costs. To address this issue, we introduce Frozen Backpropagation (\textsc{fBP}), a BP-based training algorithm relaxing weight symmetry in settings with separate networks. fBP updates forward weights by computing gradients with periodically frozen feedback weights, reducing weight transports during training and minimizing synchronization overhead. To further improve transport efficiency, we propose three partial weight transport schemes of varying computational complexity, where only a subset of weights is transported at a time. We evaluate our methods on image recognition tasks using both temporally and rate-coded SNNs, and compare them to existing approaches addressing the weight symmetry requirement. Our results show that fBP outperforms these methods and achieves accuracy comparable to BP while significantly lowering transport costs. With partial weight transport, fBP can further lower those costs by up to 10,000x at the expense of moderate accuracy loss. This work provides insights for guiding the design of neuromorphic hardware incorporating BP-based on-chip learning.
Authors: Bruno Viti, Elias Karabelas, Martin Holler
Abstract: Most machine learning-based image segmentation models produce pixel-wise confidence scores that represent the model's predicted probability for each class label at every pixel. While this information can be particularly valuable in high-stakes domains such as medical imaging, these scores are heuristic in nature and do not constitute rigorous quantitative uncertainty estimates. Conformal prediction (CP) provides a principled framework for transforming heuristic confidence scores into statistically valid uncertainty estimates. However, applying CP directly to image segmentation ignores the spatial correlations between pixels, a fundamental characteristic of image data. This can result in overly conservative and less interpretable uncertainty estimates. To address this, we propose CONSIGN (Conformal Segmentation Informed by Spatial Groupings via Decomposition), a CP-based method that incorporates spatial correlations to improve uncertainty quantification in image segmentation. Our method generates meaningful prediction sets that come with user-specified, high-probability error guarantees. It is compatible with any pre-trained segmentation model capable of generating multiple sample outputs. We evaluate CONSIGN against two CP baselines across three medical imaging datasets and two COCO dataset subsets, using three different pre-trained segmentation models. Results demonstrate that accounting for spatial structure significantly improves performance across multiple metrics and enhances the quality of uncertainty estimates.
Authors: Kewei Lian, Shaofei Cai, Yitao Liang, Anji Liu
Abstract: The ability to simulate the world in a spatially consistent manner is a crucial requirement for effective world models. Such a model enables high-quality visual generation, and also ensures the reliability of world models for downstream tasks such as simulation and planning. It must not only retain long-horizon observational information, but also enables the construction of explicit or implicit internal spatial representations. However, existing datasets do not explicitly enforce spatial consistency constraints, limiting both the ability to systematically evaluate this capability and to learn it through data-driven approaches. Furthermore, most existing benchmarks primarily emphasize visual coherence or generation quality, neglecting the requirement of long-range spatial consistency. To bridge this gap, we propose LoopNav, a dataset and corresponding benchmark centered on loop-based navigation for evaluating spatial consistency. The dataset comprises 250 hours (20 million frames) of loop-based navigation videos with actions, collected from diverse locations in the open-world environment of Minecraft. We further introduce a Scene Graph Consistency Score to quantify spatial consistency while remaining invariant to pixel-level variations. Dataset, benchmark, and code are open-sourced to support future research.
Authors: Tian Xia, Fabio De Sousa Ribeiro, Rajat R Rasal, Avinash Kori, Raghav Mehta, Ben Glocker
Abstract: Counterfactual generation aims to simulate realistic hypothetical outcomes under causal interventions. Diffusion models have emerged as a powerful tool for this task, combining DDIM inversion with conditional generation and classifier-free guidance (CFG). In this work, we identify a key limitation of CFG for counterfactual generation: it prescribes a global guidance scale for all attributes, leading to significant spurious changes in inferred counterfactuals. To mitigate this, we propose Factored Classifier-Free Guidance (FCFG), a flexible and model-agnostic guidance technique that enables attribute-wise control following a causal graph. FCFG complements recent advances in classifier-free guidance and can be seamlessly extended to advanced guidance schemes such as CFG++ and APG. Our experiments demonstrate that FCFG significantly improves the axiomatic soundness of inferred counterfactuals across both natural and medical image datasets, mitigating spurious amplification effects, and enhancing counterfactual reversibility.
Authors: Jiazhen Yan, Fan Wang, Weiwei Jiang, Ziqiang Li, Zhangjie Fu
Abstract: The rapid progress of generative models, such as GANs and diffusion models, has facilitated the creation of highly realistic images, raising growing concerns over their misuse in security-sensitive domains. While existing detectors perform well under known generative settings, they often fail to generalize to unknown generative models, especially when semantic content between real and fake images is closely aligned. In this paper, we revisit the use of CLIP features for AI-generated image detection and uncover a critical limitation: the high-level semantic information embedded in CLIP's visual features hinders effective discrimination. To address this, we propose NS-Net, a novel detection framework that leverages NULL-Space projection to decouple semantic information from CLIP's visual features, followed by contrastive learning to capture intrinsic distributional differences between real and generated images. Furthermore, we design a Patch Selection strategy to preserve fine-grained artifacts by mitigating semantic bias caused by global image structures. Extensive experiments on an open-world benchmark comprising images generated by 40 diverse generative models show that NS-Net outperforms existing state-of-the-art methods, achieving a 7.4\% improvement in detection accuracy, thereby demonstrating strong generalization across both GAN- and diffusion-based image generation techniques.
Authors: Rujosh Polma, Krishnan Menon Iyer
Abstract: As the application of deep learning in dermatology continues to grow, the recognition of melanoma has garnered significant attention, demonstrating potential for improving diagnostic accuracy. Despite advancements in image classification techniques, existing models still face challenges in identifying subtle visual cues that differentiate melanoma from benign lesions. This paper presents a novel Deeply Dual Supervised Learning framework that integrates local and global feature extraction to enhance melanoma recognition. By employing a dual-pathway structure, the model focuses on both fine-grained local features and broader contextual information, ensuring a comprehensive understanding of the image content. The framework utilizes a dual attention mechanism that dynamically emphasizes critical features, thereby reducing the risk of overlooking subtle characteristics of melanoma. Additionally, we introduce a multi-scale feature aggregation strategy to ensure robust performance across varying image resolutions. Extensive experiments on benchmark datasets demonstrate that our framework significantly outperforms state-of-the-art methods in melanoma detection, achieving higher accuracy and better resilience against false positives. This work lays the foundation for future research in automated skin cancer recognition and highlights the effectiveness of dual supervised learning in medical image analysis.
Authors: Yuru Xiao, Zihan Lin, Chao Lu, Deming Zhai, Kui Jiang, Wenbo Zhao, Wei Zhang, Junjun Jiang, Huanran Wang, Xianming Liu
Abstract: Dynamic urban scene modeling is a rapidly evolving area with broad applications. While current approaches leveraging neural radiance fields or Gaussian Splatting have achieved fine-grained reconstruction and high-fidelity novel view synthesis, they still face significant limitations. These often stem from a dependence on pre-calibrated object tracks or difficulties in accurately modeling fast-moving objects from undersampled capture, particularly due to challenges in handling temporal discontinuities. To overcome these issues, we propose a novel video diffusion-enhanced 4D Gaussian Splatting framework. Our key insight is to distill robust, temporally consistent priors from a test-time adapted video diffusion model. To ensure precise pose alignment and effective integration of this denoised content, we introduce two core innovations: a joint timestamp optimization strategy that refines interpolated frame poses, and an uncertainty distillation method that adaptively extracts target content while preserving well-reconstructed regions. Extensive experiments demonstrate that our method significantly enhances dynamic modeling, especially for fast-moving objects, achieving an approximate PSNR gain of 2 dB for novel view synthesis over baseline approaches.
Authors: Aarav Mehta, Priya Deshmukh, Vikram Singh, Siddharth Malhotra, Krishnan Menon Iyer, Tanvi Iyer
Abstract: Accurate localization of organ boundaries is critical in medical imaging for segmentation, registration, surgical planning, and radiotherapy. While deep convolutional networks (ConvNets) have advanced general-purpose edge detection to near-human performance on natural images, their outputs often lack precise localization, a limitation that is particularly harmful in medical applications where millimeter-level accuracy is required. Building on a systematic analysis of ConvNet edge outputs, we propose a medically focused crisp edge detector that adapts a novel top-down backward refinement architecture to medical images (2D and volumetric). Our method progressively upsamples and fuses high-level semantic features with fine-grained low-level cues through a backward refinement pathway, producing high-resolution, well-localized organ boundaries. We further extend the design to handle anisotropic volumes by combining 2D slice-wise refinement with light 3D context aggregation to retain computational efficiency. Evaluations on several CT and MRI organ datasets demonstrate substantially improved boundary localization under strict criteria (boundary F-measure, Hausdorff distance) compared to baseline ConvNet detectors and contemporary medical edge/contour methods. Importantly, integrating our crisp edge maps into downstream pipelines yields consistent gains in organ segmentation (higher Dice scores, lower boundary errors), more accurate image registration, and improved delineation of lesions near organ interfaces. The proposed approach produces clinically valuable, crisp organ edges that materially enhance common medical-imaging tasks.
Authors: Vikram Singh, Kabir Malhotra, Rohan Desai, Ananya Shankaracharya, Priyadarshini Chatterjee, Krishnan Menon Iyer
Abstract: Lesion segmentation, in contrast to natural scene segmentation, requires handling subtle variations in texture and color, frequent imaging artifacts (such as hairs, rulers, and bubbles), and a critical need for precise boundary localization to aid in accurate diagnosis. The accurate delineation of melanocytic tumors in dermoscopic images is a crucial component of automated skin cancer screening systems and clinical decision support. In this paper, we present a novel dual-resolution architecture inspired by ResNet, specifically tailored for the segmentation of melanocytic tumors. Our approach incorporates a high-resolution stream that preserves fine boundary details, alongside a complementary pooled stream that captures multi-scale contextual information for robust lesion recognition. These two streams are closely integrated through boundary-aware residual connections, which inject edge information into deep feature maps, and a channel attention mechanism that adapts the model's sensitivity to color and texture variations in dermoscopic images. To tackle common imaging artifacts and the challenges posed by small clinical datasets, we introduce a lightweight artifact suppression block and a multi-task training strategy. This strategy combines the Dice-Tversky loss with an explicit boundary loss and a contrastive regularizer to enhance feature stability. This unified design enables the model to generate pixel-accurate segmentation masks without the need for extensive post-processing or complex pre-training. Extensive evaluation on public dermoscopic benchmarks reveals that our method significantly enhances boundary precision and clinically relevant segmentation metrics, outperforming traditional encoder-decoder baselines. This makes our approach a valuable component for building automated melanoma assessment systems.
Authors: Ayaan Nooruddin Siddiqui, Mahnoor Zaidi, Ayesha Nazneen Shahbaz, Priyadarshini Chatterjee, Krishnan Menon Iyer
Abstract: The task of parsing subcutaneous vessels in clinical images is often hindered by the high cost and limited availability of ground truth data, as well as the challenge of low contrast and noisy vessel appearances across different patients and imaging modalities. In this work, we propose a novel weakly supervised training framework specifically designed for subcutaneous vessel segmentation. This method utilizes low-cost, sparse annotations such as centerline traces, dot markers, or short scribbles to guide the learning process. These sparse annotations are expanded into dense probabilistic supervision through a differentiable random walk label propagation model, which integrates vesselness cues and tubular continuity priors driven by image data. The label propagation process results in per-pixel hitting probabilities and uncertainty estimates, which are incorporated into an uncertainty-weighted loss function to prevent overfitting in ambiguous areas. Notably, the label propagation model is trained jointly with a CNN-based segmentation network, allowing the system to learn vessel boundaries and continuity constraints without the need for explicit edge supervision. Additionally, we introduce a topology-aware regularizer that encourages centerline connectivity and penalizes irrelevant branches, further enhancing clinical applicability. Our experiments on clinical subcutaneous imaging datasets demonstrate that our approach consistently outperforms both naive sparse-label training and traditional dense pseudo-labeling methods, yielding more accurate vascular maps and better-calibrated uncertainty, which is crucial for clinical decision-making. This method significantly reduces the annotation workload while maintaining clinically relevant vessel topology.
Authors: Haoyue Li, Yifan Gao, Feng Yuan, Xiaosong Wang, Xin Gao
Abstract: Foundation models pre-trained on large-scale natural image datasets offer a powerful paradigm for medical image segmentation. However, effectively transferring their learned representations for precise clinical applications remains a challenge. In this work, we propose Dino U-Net, a novel encoder-decoder architecture designed to exploit the high-fidelity dense features of the DINOv3 vision foundation model. Our architecture introduces an encoder built upon a frozen DINOv3 backbone, which employs a specialized adapter to fuse the model's rich semantic features with low-level spatial details. To preserve the quality of these representations during dimensionality reduction, we design a new fidelity-aware projection module (FAPM) that effectively refines and projects the features for the decoder. We conducted extensive experiments on seven diverse public medical image segmentation datasets. Our results show that Dino U-Net achieves state-of-the-art performance, consistently outperforming previous methods across various imaging modalities. Our framework proves to be highly scalable, with segmentation accuracy consistently improving as the backbone model size increases up to the 7-billion-parameter variant. The findings demonstrate that leveraging the superior, dense-pretrained features from a general-purpose foundation model provides a highly effective and parameter-efficient approach to advance the accuracy of medical image segmentation. The code is available at https://github.com/yifangao112/DinoUNet.
Authors: Yehudit Aperstein, Alexander Apartsin
Abstract: Early-exit cascades over a frozen convolutional backbone enable adaptive inference but suffer from three sources of train-inference mismatch: branches train on samples they will never see at inference, their per-class precision thresholds are calibrated on the wrong distribution, and the standard cross-entropy target on backbone argmax labels discards the backbone's uncertainty signal. We close all three gaps with CalexNet (Cascade-Aligned Early eXits), a training-recipe-only modification: branches train under continuously-weighted importance sampling that matches the cascade-survivor distribution; per-class precision thresholds are calibrated on the actual cascade-survivor subset of the validation set; and the classification head is trained against the backbone's full softmax via a temperature-scaled KL objective. Combined with an augmented prototype-pooling branch head, CalexNet is evaluated on ResNet18 and ResNet50 backbones across CIFAR-100 (20-superclass coarse, the harder primary setting) and CINIC-10 (10-class, the easier cross-validation counterpart). On the accuracy-FLOPs Pareto frontier, CalexNet matches or exceeds three published baselines (PTEEnet, ZTW, BoostNet) and a within-paper "no-alignment, no-KD" reference. The largest gains appear in the practically relevant 30-70% FLOPs-reduction regime and are stable across n=3 training seeds. CalexNet requires no inference-time architectural change and is a drop-in for any frozen-backbone early-exit cascade.
Authors: Haiyu Yang, Enhong Liu, Jennifer Sun, Sumit Sharma, Meike van Leerdam, Sebastien Franceschini, Puchun Niu, Miel Hostens
Abstract: Animal behavior analysis plays a crucial role in understanding animal welfare, health status, and productivity in agricultural settings. However, traditional manual observation methods are time-consuming, subjective, and limited in scalability. We present a modular pipeline that leverages open-sourced state-of-the-art computer vision techniques to automate animal behavior analysis in a group housing environment. Our approach combines state-of-the-art models for zero-shot object detection, motion-aware segmentation and tracking, and advanced feature extraction using vision transformers for robust behavior recognition. The pipeline addresses challenges including animal occlusions and group housing scenarios, as demonstrated in indoor pig monitoring. We validated our system on the Edinburgh Pig Behavior Video Dataset for multiple behavioral tasks. Our temporal model achieved 94.2% overall accuracy, representing a 21.2 percentage point improvement over existing methods. The pipeline demonstrated robust tracking capabilities with a 93.3% identity preservation (IDF1) score and an 89.3% average precision (AP) for object detection. The modular design suggests potential for adaptation to other contexts, though further validation across species would be required. The open-source implementation provides a scalable solution for behavior monitoring, contributing to precision pig farming and welfare assessment through automated, objective, and continuous analysis.
Authors: Sahar Dastani, Ali Bahri, Gustavo Adolfo Vargas Hakim, Moslem Yazdanpanah, Mehrdad Noori, David Osowiechi, Samuel Barbeau, Ismail Ben Ayed, Herve Lombaert, Christian Desrosiers
Abstract: State Space Models (SSMs) have emerged as efficient alternatives to Vision Transformers (ViTs), with VMamba standing out as a pioneering architecture designed for vision tasks. However, their generalization performance degrades significantly under distribution shifts. To address this limitation, we propose TRUST (Test-Time Refinement using Uncertainty-Guided SSM Traverses), a novel test-time adaptation (TTA) method that leverages diverse traversal permutations to generate multiple causal perspectives of the input image. Model predictions serve as pseudo-labels to guide updates of the Mamba-specific parameters, and the adapted weights are averaged to integrate the learned information across traversal scans. Altogether, TRUST is the first approach that explicitly leverages the unique architectural properties of SSMs for adaptation. Experiments on seven benchmarks show that TRUST consistently improves robustness and outperforms existing TTA methods.
Authors: Zhaohua Zhang, Jianhuan Zhuo, Muxi Chen, Chenchen Zhao, Wenyu Jiang, Tianwen Jiang, Mingyang Chen, Yutang, Qiuyong Xiao, Jihong Zhang, Zhixun Su
Abstract: The CLIP model has established itself as a cornerstone of large-scale retrieval systems. However, its performance often degrades under distributional shifts such as multilingual, long-form, or multimodal queries. To avoid the prohibitive costs associated with retriever retraining or corpus re-embedding, we propose GRAPE (Grouped Ranking-Aware Policy Optimization Enhancement), a plug-and-play approach that leverages LLM-based query rewriting to bridge these gaps. Unlike existing methods that lack explicit supervision, GRAPE integrates ranking signals into the rewriting LLM via Grouped Relative Policy Optimization (GRPO), ensuring rewritten queries are better aligned with the frozen retriever's latent distribution. Crucially, we identify a score inflation phenomenon in naive similarity-based finetuning - where irrelevant candidates receive indiscriminately high scores - and mitigate it with a novel corpus-relative ranking-based reward. Extensive experiments across multilingual (Flickr30k-CN, CVLUE, XM3600), long-form (Wikipedia), and multimodal (CIRR) benchmarks demonstrate that GRAPE consistently improves performance, achieving an average gain of 4.9% in Recall@10 without any modification to the underlying retriever. The code is available at https://github.com/mogulzhang/GRAPE.
Authors: Tuan Nguyen, Naseem Khan, Khang Tran, NhatHai Phan, Issa Khalil
Abstract: The rapid rise of synthetic media has made deepfake detection a critical challenge for online safety and trust. Progress remains constrained by the scarcity of large, high-quality datasets. Although multimodal large language models (LLMs) exhibit strong reasoning capabilities, their performance on deepfake detection is poor, often producing explanations that are misaligned with visual evidence or hallucinatory. To address this limitation, we introduce a reasoning-annotated dataset for deepfake detection and propose Paragraph-level Relative Policy Optimization (PRPO), a reinforcement learning algorithm that aligns LLM reasoning with image content at the paragraph level. Experiments show that PRPO improves detection accuracy by a wide margin and achieves the highest reasoning score of 4.55/5.0. Ablation studies further demonstrate that PRPO significantly outperforms GRPO under test-time conditions. These results underscore the importance of grounding multimodal reasoning in visual evidence to enable more reliable and interpretable deepfake detection.
Authors: Thomas Fel, Binxu Wang, Michael A. Lepori, Matthew Kowal, Andrew Lee, Randall Balestriero, Sonia Joseph, Ekdeep S. Lubana, Talia Konkle, Demba Ba, Martin Wattenberg
Abstract: DINOv2 is routinely deployed to recognize objects, scenes, and actions; yet the nature of what it perceives remains unknown. As a working baseline, we adopt the Linear Representation Hypothesis (LRH) and operationalize it using SAEs, producing a 32,000-unit dictionary that serves as the interpretability backbone of our study, which unfolds in three parts. In the first part, we analyze how different downstream tasks recruit concepts from our learned dictionary, revealing functional specialization: classification exploits "Elsewhere" concepts that fire everywhere except on target objects, implementing learned negations; segmentation relies on boundary detectors forming coherent subspaces; depth estimation draws on three distinct monocular depth cues matching visual neuroscience principles. Following these functional results, we analyze the geometry and statistics of the concepts learned by the SAE. We found that representations are partly dense rather than strictly sparse. The dictionary evolves toward greater coherence and departs from maximally orthogonal ideals (Grassmannian frames). Within an image, tokens occupy a low dimensional, locally connected set persisting after removing position. These signs suggest representations are organized beyond linear sparsity alone. Synthesizing these observations, we propose a refined view: tokens are formed by combining convex mixtures of archetypes (e.g., a rabbit among animals, brown among colors, fluffy among textures). This structure is grounded in Gardenfors' conceptual spaces and in the model's mechanism as multi-head attention produces sums of convex mixtures, defining regions bounded by archetypes. We introduce the Minkowski Representation Hypothesis (MRH) and examine its empirical signatures and implications for interpreting vision-transformer representations.
Authors: Rui-Yang Ju, Kohei Yamashita, Hirotaka Kameko, Shinsuke Mori
Abstract: Kuzushiji, a pre-modern Japanese cursive script, can currently be read and understood by only a few thousand trained experts in Japan. With the rapid development of deep learning, researchers have begun applying Optical Character Recognition (OCR) techniques to transcribe Kuzushiji into modern Japanese. Although existing OCR methods perform well on clean pre-modern Japanese documents written in Kuzushiji, they often fail to consider various types of noise, such as document degradation and seals, which significantly affect recognition accuracy. To the best of our knowledge, no existing dataset specifically addresses these challenges. To address this gap, we introduce the Degraded Kuzushiji Documents with Seals (DKDS) dataset as a new benchmark for related tasks. We describe the dataset construction process, which involves the assistance of a trained Kuzushiji expert, and define two benchmark tracks: (1) Kuzushiji character and seal detection and (2) document binarization. For the Kuzushiji character and seal detection track, we provide baseline results using several recent versions of YOLO to detect Kuzushiji characters and seals. For the document binarization track, we present baseline results from traditional binarization algorithms, traditional algorithms combined with K-means clustering, two state-of-the-art (SOTA) generative adversarial network (GAN) methods, and our improved conditional GAN (cGAN)-based method. The DKDS dataset and the implementation code for baseline methods are available at https://ruiyangju.github.io/DKDS.
Authors: Shengqin Jiang, Tianqi Kong, Yuankai Qi, Haokui Zhang, Lina Yao, Quan Z. Sheng, Qingshan Liu, Ming-Hsuan Yang
Abstract: Prompt-based continual learning methods fine-tune only a small set of additional learnable parameters while keeping the pre-trained model's parameters frozen. It enables efficient adaptation to new tasks while mitigating the risk of catastrophic forgetting. These methods typically attach one independent task-specific prompt to each layer of pre-trained models to locally modulate its features, ensuring that the layer's representation aligns with the requirements of the new task. However, although introducing learnable prompts independently at each layer provides high flexibility for adapting to new tasks, this overly flexible tuning could make certain layers susceptible to unnecessary updates. As all prompts till the current task are added together as a final prompt for all seen tasks, the model may easily overwrite feature representations essential to previous tasks, which increases the risk of catastrophic forgetting. To address this issue, we propose a novel hierarchical layer-grouped prompt tuning method for continual learning. It improves model stability in two ways: (i) Layers in the same group share roughly the same prompts, which are adjusted by position encoding. This helps preserve the intrinsic feature relationships and propagation pathways of the pre-trained model within each group. (ii) It utilizes a single task-specific root prompt to learn to generate sub-prompts for each layer group. In this way, all sub-prompts are conditioned on the same root prompt, enhancing their synergy and reducing independence. Extensive experiments across four benchmarks demonstrate that our method achieves favorable performance compared with several state-of-the-art methods.
Authors: Kishor Datta Gupta, Marufa Kamal, Md. Mahfuzur Rahman, Fahad Rahman, Mohd Ariful Haque, Sunzida Siddique
Abstract: Current state of the art measures like BLEU, CIDEr, VQA score, SigLIP-2 and CLIPScore are often unable to capture semantic or structural accuracy, especially for domain-specific or context-dependent scenarios. For this, this paper proposes a Physics-Constrained Multimodal Data Evaluation (PCMDE) metric combining large language models with reasoning, knowledge based mapping and vision-language models to overcome these limitations. The architecture is comprised of three main stages: (1) feature extraction of spatial and semantic information with multimodal features through object detection and VLMs; (2) Confidence-Weighted Component Fusion for adaptive component-level validation; and (3) physics-guided reasoning using large language models for structural and relational constraints (e.g., alignment, position, consistency) enforcement.
Authors: Md Sohag Mia, Md Nahid Hasan, Muhammad Abdullah Adnan
Abstract: Despite significant progress in 3D object detection, point clouds remain challenging due to sparse data, incomplete structures, and limited semantic information. Capturing contextual relationships between distant objects presents additional difficulties. To address these challenges, we propose GraphFusion3D, a unified framework combining multi-modal fusion with advanced feature learning. Our approach introduces the Adaptive Cross-Modal Transformer (ACMT), which adaptively integrates image features into point representations to enrich both geometric and semantic information. For proposal refinement, we introduce the Graph Reasoning Module (GRM), a novel mechanism that models neighborhood relationships to simultaneously capture local geometric structures and global semantic context. The module employs multi-scale graph attention to dynamically weight both spatial proximity and feature similarity between proposals. We further employ a cascade decoder that progressively refines detections through multi-stage predictions. Extensive experiments on SUN RGB-D (70.6% AP$_{25}$ and 51.2% AP$_{50}$) and ScanNetV2 (75.1% AP$_{25}$ and 60.8% AP$_{50}$) demonstrate a substantial performance improvement over existing approaches.
Authors: Haicheng Liao, Huanming Shen, Bonan Wang, Yongkang Li, Yihong Tang, Chengyue Wang, Dingyi Zhuang, Kehua Chen, Hai Yang, Chengzhong Xu, Zhenning Li
Abstract: Interpreting natural-language commands to localize target objects is critical for autonomous driving (AD). Existing visual grounding (VG) methods for autonomous vehicles (AVs) typically struggle with ambiguous, context-dependent instructions, as they lack reasoning over 3D spatial relations and anticipated scene evolution. Grounded in the principles of world models, we propose ThinkDeeper, a framework that reasons about future spatial states before making grounding decisions. At its core is a Spatial-Aware World Model (SA-WM) that learns to reason ahead by distilling the current scene into a command-aware latent state and rolling out a sequence of future latent states, providing forward-looking cues for disambiguation. Complementing this, a hypergraph-guided decoder then hierarchically fuses these states with the multimodal input, capturing higher-order spatial dependencies for robust localization. In addition, we present DrivePilot, a multi-source VG dataset in AD, featuring semantic annotations generated by a Retrieval-Augmented Generation (RAG) and Chain-of-Thought (CoT)-prompted LLM pipeline. Extensive evaluations on six benchmarks, ThinkDeeper ranks #1 on the Talk2Car leaderboard and surpasses state-of-the-art baselines on DrivePilot, MoCAD, and RefCOCO/+/g benchmarks. Notably, it shows strong robustness and efficiency in challenging scenes (long-text, multi-agent, ambiguity) and retains superior performance even when trained on 50% of the data.
Authors: Wenliang Guo, Yu Kong
Abstract: Procedural activities are fundamentally driven by object state transitions, yet existing instructional video benchmarks remain action-centric and cannot evaluate whether models reason about how objects evolve toward task completion. In this work, we introduce ProcObject-10K, the first benchmark that jointly evaluates object-centric reasoning and temporal evidence grounding in instructional videos, across both egocentric and exocentric views. It comprises 10,522 open-ended VideoQA pairs grounded in 1,799 video clips, spanning 137 tasks across 9 domains and five reasoning types covering preconditions, state evolution, counterfactuals, mistakes, and readiness. Benchmarking 13 leading MLLMs reveals a substantial answering-grounding gap: models produce plausible answers while failing to localize the supporting evidence (mIoU < 45%), exposing their reliance on linguistic priors rather than fine-grained object dynamics. As a step toward closing this gap, we further provide an object-centric supervised fine-tuning baseline with pseudo object-level supervision and spatial-temporal constraints. Models fine-tuned on ProcObject-10K not only improve on the benchmark itself, but also transfer effectively to other grounded VideoQA and embodied planning tasks. The dataset, annotations, and evaluation toolkit will be publicly released to support future research on object-centric procedural understanding.
Authors: Md. Sanaullah Chowdhury Lameya Sabrin
Abstract: Medical image segmentation requires balancing local precision for boundary-critical clinical applications, global context for anatomical coherence, and computational efficiency for deployment on limited data and hardware a trilemma that existing architectures fail to resolve. Although convolutional networks provide local precision at $\mathcal{O}(n)$ cost but limited receptive fields, vision transformers achieve global context through $\mathcal{O}(n^2)$ self-attention at prohibitive computational expense, causing overfitting on small clinical datasets. We propose S2M-Net, a 4.7M-parameter architecture that achieves $\mathcal{O}(HW \log HW)$ global context through two synergistic innovations: (i) Spectral-Selective Token Mixer (SSTM), which exploits the spectral concentration of medical images via truncated 2D FFT with learnable frequency filtering and content-gated spatial projection, avoiding quadratic attention cost while maintaining global receptive fields; and (ii) Morphology-Aware Adaptive Segmentation Loss (MASL), which automatically analyzes structure characteristics (compactness, tubularity, irregularity, scale) to modulate five complementary loss components through constrained learnable weights, eliminating manual per-dataset tuning. Comprehensive evaluation in 16 medical imaging datasets that span 8 modalities demonstrates state-of-the-art performance: 96.12\% Dice on polyp segmentation, 83.77\% on surgical instruments (+17.85\% over the prior art) and 80.90\% on brain tumors, with consistent 3-18\% improvements over specialized baselines while using 3.5--6$\times$ fewer parameters than transformer-based methods.
Authors: Zhipeng Qian, Zihan Liang, Yufei Ma, Ben Chen, Huangyu Dai, Yiwei Ma, Jiayi Ji, Chenyi Lei, Han Li, Xiaoshuai Sun
Abstract: Composed Image Retrieval (CIR) enables users to search for target images using both a reference image and manipulation text, offering substantial advantages over single-modality retrieval systems. However, existing CIR methods suffer from representation space fragmentation: queries and targets comprise heterogeneous modalities and are processed by distinct encoders, forcing models to bridge misaligned representation spaces only through post-hoc alignment, which fundamentally limits retrieval performance. This architectural asymmetry manifests as three distinct, well-separated clusters in the feature space, directly demonstrating how heterogeneous modalities create fundamentally misaligned representation spaces from initialization. In this work, we propose CSMCIR, a unified representation framework that achieves efficient query-target alignment through three synergistic components. First, we introduce a Multi-level Chain-of-Thought (MCoT) prompting strategy that guides Multimodal Large Language Models to generate discriminative, semantically compatible captions for target images, establishing modal symmetry. Building upon this, we design a symmetric dual-tower architecture where both query and target sides utilize the identical shared-parameter Q-Former for cross-modal encoding, ensuring consistent feature representations and further reducing the alignment gap. Finally, this architectural symmetry enables an entropy-based, temporally dynamic Memory Bank strategy that provides high-quality negative samples while maintaining consistency with the evolving model state. Extensive experiments on four benchmark datasets demonstrate that our CSMCIR achieves state-of-the-art performance with superior training efficiency. Comprehensive ablation studies further validate the effectiveness of each proposed component.
Authors: Jinrui Yang, Qing Liu, Yijun Li, Mengwei Ren, Letian Zhang, Zhe Lin, Cihang Xie, Yuyin Zhou
Abstract: Recent image generation models produce impressive composites, but often fail to preserve the identity of user-provided content when editing specific elements: the surrounding scene may shift, and even the edited object's appearance can drift from the original. Layered representation offer a natural remedy--they allow users to independently manipulate individual elements--but existing layered methods typically produce transparent foregrounds without realistic visual effects such as shadows and reflections, forcing the use of a second harmonization model after every edit, which in turn introduces drift. To overcome these limitations, we present LASAGNA, which generates a photorealistic background (BG) and an RGBA foreground with compelling visual effects in a single forward pass. By treating object-associated visual effects as part of the foreground (FG) layer, LASAGNA supports the dominant class of consumer edits (e.g., translation, scaling, recoloring, duplication) via alpha compositing alone, without invoking any model post-edit, thereby eliminating identity drift introduced by cascade editing pipelines. This single-pass design contrasts with prior layered methods that rely on separate expert models for each task. LASAGNA handles diverse conditional inputs--text prompts, FG, BG, and location masks--within a unified architecture. We further release two community resources: LASAGNA-48K, the first public dataset of 48K layered image triplets with photorealistic visual effects, and LASAGNA-BENCH, the first standardized benchmark for layer-centric generation and editing, comprising 242 expert-annotated samples across six diverse sources. Experiments show that LASAGNA outperforms both general-purpose editors and prior layered methods across three generation modes, and supports a wide range of post-edits without any model re-inference.
Authors: Yifan Chen, Fei Yin, Hao Chen, Jia Wu, Chao Li
Abstract: Contrast-enhanced imaging is central to oncologic diagnosis, but contrast agents can be contraindicated for many of the patients who need them most. Synthesizing contrast scans from non-contrast inputs is the natural response. Two obstacles stand in the way: no benchmark provides paired contrast data with lesion-level evaluation, and no single model handles the arbitrary missing patterns seen in practice. We introduce Contrast-X, a benchmark of paired contrast-enhanced and non-contrast imaging spanning 10 organs in CT (1{,}526 patients) and multi-phase breast DCE-MRI (1116 patients). Every case carries radiologist-verified phase labels and tumor masks. We further propose FlowMI, a single model that handles arbitrary subsets of available modalities through a unified multi-modal latent space and flow matching. We benchmark a range of missing-modality configurations, reporting standard image-quality metrics, radiologist reader studies, and downstream lesion analysis on the synthesized scans. We further evaluate cross-organ generalization to test whether the model has learned a transferable contrast-enhancement operation. Dataset, code, and leaderboard will be released. Our code are available at https://github.com/YifanChen02/Contrast-X.
Authors: Renjie Ding, Yaonan Wang, Min Liu, Jialin Zhu, Jiazheng Wang, Jiahao Zhao, Wenting Shen, Feixiang He, Xiang Chen
Abstract: 3D Gaussian Splatting (3DGS) has emerged as a powerful technique for real-time novel view synthesis. As an explicit representation optimized through gradient propagation among primitives, optimization widely accepted in deep neural networks (DNNs) is actually adopted in 3DGS, such as synchronous weight updating and Adam with the adaptive gradient. However, considering the physical significance and specific design in 3DGS, there are two overlooked details in the optimization of 3DGS: (i) update step coupling, which induces optimizer state rescaling and costly attribute updates outside the viewpoints, and (ii) gradient coupling in the moment, which may lead to under- or over-effective regularization. Nevertheless, such a complex coupling is under-explored. After revisiting the optimization of 3DGS, we take a step to decouple it and recompose the process into: Sparse Adam, Re-State Regularization and Decoupled Attribute Regularization. Taking a large number of experiments under the 3DGS and 3DGS-MCMC frameworks, our work provides a deeper understanding of these components. Finally, based on the empirical analysis, we re-design the optimization and propose AdamW-GS by re-coupling the beneficial components, under which better optimization efficiency and representation effectiveness are achieved simultaneously.
Authors: Bishoy Galoaa, Xiangyu Bai, Sarah Ostadabbas
Abstract: State-of-the-art vision-language models (VLMs) score impressively on video benchmarks yet stumble on basic visual reasoning tasks involving spatial relations, navigation, and object selection that a preschooler solves easily. We hypothesize that the explicit pedagogical structure, specifically the context-question-pause-answer cycles embedded in children's educational video, provides naturally co-aligned reasoning traces: temporally synchronized visual cues, questions, and answers that emerge only from deliberate pedagogical authoring and cannot be practically reconstructed through manual annotation at scale. To test this, we introduce SoSVQA (Structure over Scale Visual Question Answering), a unified benchmark of 10K question-answer pairs automatically extracted from Dora the Explorer (DoraVQA) and Mickey Mouse Clubhouse (ClubHVQA) with precise timestamp alignment, and fine-tune Qwen2-VL and Qwen3-VL using Group Relative Policy Optimization (GRPO) to leverage the clear correctness signals and structured reasoning traces inherent in educational content. Despite training on just 10K QA pairs from 78 hours of children's television, orders of magnitude less data than GPT and Gemini, our approach delivers generalizable performance gains for Qwen-based VLMs, yielding consistent improvements on NExT-QA (+19.7), Video-MME (+10.6), and MotionBench (+4.9), matching the performance of leading proprietary systems and demonstrating that content structure can compensate for content scale.
Authors: Jin-Chuan Shi, Binhong Ye, Tao Liu, Xiaoyang Liu, Yangjinhui Xu, Junzhe He, Zeju Li, Hao Chen, Chunhua Shen
Abstract: Reconstructing dynamic hand-object interactions from monocular videos is critical for dexterous manipulation data collection and creating realistic digital twins for robotics and VR. However, current methods face two prohibitive barriers: (1) reliance on neural rendering often yields fragmented, non-simulation-ready geometries under heavy occlusion, and (2) dependence on brittle Structure-from-Motion (SfM) initialization leads to frequent failures on in-the-wild footage. To overcome these limitations, we introduce AGILE, a robust framework that shifts the paradigm from reconstruction to agentic generation for interaction learning. First, we employ an agentic pipeline where a Vision-Language Model (VLM) guides a generative model to synthesize a complete, watertight object mesh with high-fidelity texture, independent of video occlusions. Second, bypassing fragile SfM entirely, we propose a robust anchor-and-track strategy. We initialize the object pose at a single interaction onset frame using a foundation model and propagate it temporally by leveraging the strong visual similarity between our generated asset and video observations. Finally, a contact-aware optimization integrates semantic, geometric, and interaction stability constraints to enforce physical plausibility. Extensive experiments on HO3D, DexYCB, ARCTIC, and in-the-wild videos reveal that AGILE outperforms baselines in global geometric accuracy while demonstrating exceptional robustness on challenging sequences where prior arts frequently collapse. By prioritizing physical validity, our method produces simulation-ready assets validated via real-to-sim retargeting for robotic applications. Project page: https://agile-hoi.github.io.
Authors: Roberto Leotta, Salvatore Alfio Sambataro, Claudio Vittorio Ragaglia, Mirko Casu, Yuri Petralia, Francesco Guarnera, Luca Guarnera, Sebastiano Battiato
Abstract: Modern T2V/I2V generators synthesize people increasingly hard to distinguish from authentic footage, while current evaluation suites lag: legacy benchmarks target manipulation-based forgeries, and recent synthetic-video benchmarks prioritize scale over realistic human depiction. We introduce SynthForensics, a people-centric benchmark of $20{,}445$ videos from 8 T2V and 7 I2V open-source generators, paired-source from FF++/DFD reals, two-stage human-validated, in four compression versions with full metadata. In our paired-comparison human study, raters prefer SynthForensics in $71$--$77\%$ of head-to-head comparisons against each of nine existing synthetic-video benchmarks, while facial-quality metrics fall within the FF++/DFD baseline range. Across 15 detectors and three protocols, face-based methods drop $13$--$55$ AUC points (mean $27$) from FF++ to SynthForensics and a further $23$ under aggressive compression; fine-tuning closes the gap at a backward cost on legacy benchmarks; training from scratch shows synthetic and manipulation features largely disjoint for most detectors. We release dataset, pipeline, and code.
Authors: Yiming Zhang, Qiangyu Yan, Borui Jiang, Kai Han
Abstract: The advancement of multimodal large language models (MLLMs) has enabled impressive perception capabilities. However, their reasoning process often remains a "fast thinking" paradigm, reliant on end-to-end generation or explicit, language-centric chains of thought (CoT), which can be inefficient, verbose, and prone to hallucination. This work posits that robust reasoning should evolve within a latent space, integrating multimodal signals seamlessly. We propose multimodal latent reasoning via HIerarchical Visual cuEs injection (\emph{HIVE}), a novel framework that instills deliberate, "slow thinking" without depending on superficial textual rationales. Our method recursively extends transformer blocks, creating an internal loop for iterative reasoning refinement. Crucially, it injectively grounds this process with hierarchical visual cues from global scene context to fine-grained regional details directly into the model's latent representations. This enables the model to perform grounded, multi-step inference entirely in the aligned latent space. Extensive evaluations demonstrate that test-time scaling is effective when incorporating vision knowledge, and that integrating hierarchical information significantly enhances the model's understanding of complex scenes.
Authors: Mridankan Mandal
Abstract: Human Activity Recognition (HAR) on resource constrained wearables requires models that balance accuracy against strict memory and computational budgets. State of the art lightweight architectures such as TinierHAR (34K parameters), and TinyHAR (55K parameters) achieve strong accuracy, but exceed memory budgets of microcontrollers with limited SRAM once operating system overhead is considered. We present MicroBi-ConvLSTM, an ultra-lightweight convolutional recurrent architecture achieving 11.4K parameters on average through two stage convolutional feature extraction with 4x temporal pooling, and a single bidirectional LSTM layer. This represents 2.9x parameter reduction versus TinierHAR, and 11.9x versus DeepConvLSTM while preserving linear O(N) complexity. Evaluation across eight diverse HAR benchmarks shows that MicroBi-ConvLSTM maintains competitive performance within the ultra-lightweight regime: 93.41% macro F1 on UCI-HAR, 94.46% on SKODA assembly gestures, and 88.98% on Daphnet gait freeze detection. Systematic ablation reveals task dependent component contributions where bidirectionality benefits episodic event detection, but provides marginal gains on periodic locomotion. On-device deployment on the Raspberry Pi Pico 2 and ESP32 validates hardware viability: MicroBi-ConvLSTM is the only architecture achieving full 8/8 dataset coverage on both platforms, with 72.8 ms average latency on Pico 2 and 97.9% PyTorch parity on ESP32, while all three baselines show partial or complete deployment failure.
Authors: Xiaomin Yu, Yi Xin, Yuhui Zhang, Wenjie Zhang, Chonghan Liu, Hanzhen Zhao, Chen Liu, Xiaoxing Hu, Ziyue Qiao, Hao Tang, Xiaobin Hu, Chengwei Qin, Hui Xiong, Yu Qiao, Shuicheng Yan
Abstract: Despite the success of multimodal contrastive learning in aligning visual and linguistic representations, a persistent geometric anomaly, the Modality Gap, remains: embeddings of distinct modalities expressing identical semantics occupy systematically offset regions. Prior approaches to bridge this gap are largely limited by oversimplified isotropic assumptions, hindering their application in large-scale scenarios. In this paper, we address these limitations by precisely characterizing the geometric shape of the modality gap and leveraging it for efficient model scaling. First, we propose the Fixed-frame Modality Gap Theory, which decomposes the modality gap within a frozen reference frame into stable biases and anisotropic residuals. Guided by this precise modeling, we introduce ReAlign, a training-free modality alignment strategy. Utilizing statistics from massive unpaired data, ReAlign aligns text representation into the image representation distribution via a three-step process comprising Anchor, Trace, and Centroid Alignment, thereby explicitly rectifying geometric misalignment. Building on ReAlign, we propose ReVision, a scalable training paradigm for Multimodal Large Language Models~(MLLMs). ReVision integrates ReAlign into the pretraining stage, enabling the model to learn the distribution of visual representations from unpaired text before visual instruction tuning, without the need for large-scale, high-quality image-text pairs. Our framework demonstrates that statistically aligned unpaired data can effectively substitute for expensive image-text pairs, offering a robust path for the efficient scaling of MLLMs.
Authors: Peng Chen, Chao Huang, Yunkang Cao, Chengliang Liu, Wei Wang, Wenqiang Wang, Mingbo Yang, Li Shen, Wenqi Ren, Xiaochun Cao
Abstract: Industrial anomaly detection demands precise reasoning over fine-grained defect patterns. However, existing multimodal large language models (MLLMs), pretrained on general-domain data, often struggle to capture category-specific anomalies, thereby limiting both detection accuracy and interpretability. To address these limitations, we propose Reason-IAD, a knowledge-guided dynamic latent reasoning framework for explainable industrial anomaly detection. Reason-IAD comprises two core components. First, a retrieval-augmented knowledge module incorporates category-specific textual descriptions into the model input, enabling context-aware reasoning over domain-specific defects. Second, an entropy-driven latent reasoning mechanism conducts iterative exploration within a compact latent space using optimizable latent think tokens, guided by an entropy-based reward that encourages confident and stable predictions. Furthermore, a dynamic visual injection strategy selectively incorporates the most informative image patches into the latent sequence, directing the reasoning process toward regions critical for anomaly detection. Extensive experimental results demonstrate that Reason-IAD consistently outperforms state-of-the-art methods across multiple tasks. The code will be publicly available at https://github.com/chenpeng052/Reason-IAD.
Authors: Manfred M. Fischer, Joshua Pitts
Abstract: This paper investigates the relationship between convolutional neural network (CNN) topology and image recognition performance through a comparative study of the VGG, ResNet, and GoogLeNet architectural families. Utilizing a unified experimental framework, the study isolates the impact of depth from confounding implementation variables. A formal distinction is introduced between nominal depth ($D_{\mathrm{nom}}$), representing the physical layer count, and effective depth ($D_{\mathrm{eff}}$), an operational metric quantifying the expected number of sequential transformations. Empirical results demonstrate that architectures utilizing identity shortcuts or branching modules maintain optimization stability by decoupling $D_{\mathrm{eff}}$ from $D_{\mathrm{nom}}$. These findings suggest that effective depth serves as a superior framework for predicting scaling potential and practical trainability, ultimately indicating that architectural topology - rather than sheer layer volume - is the primary determinant of gradient health in deep learning models.
Authors: Dong Liu, Yanxuan Yu, Ben Lengerich, Ying Nian Wu
Abstract: Diffusion Transformers (DiTs) achieve state-of-the-art performance in high-fidelity image and video generation but suffer from expensive inference due to their iterative denoising structure. While prior methods accelerate sampling by caching intermediate features, they rely on static reuse schedules or coarse-grained heuristics, which often lead to temporal drift and cache misalignment that significantly degrade generation quality. We introduce \textbf{AdaCorrection}, an adaptive offset cache correction framework that maintains high generation fidelity while enabling efficient cache reuse across Transformer layers during diffusion inference. At each timestep, AdaCorrection estimates cache validity with lightweight spatio-temporal signals and adaptively blends cached and fresh activations. This correction is computed on-the-fly without additional supervision or retraining. Our approach achieves strong generation quality with minimal computational overhead, maintaining near-original FID while providing moderate acceleration. Experiments on image and video diffusion benchmarks show that AdaCorrection consistently improves generation performance.
Authors: Cem Eteke, Batuhan Tosun, Martin Piccolrovazzi, Alexander Griessel, Wolfgang Kellerer, Eckehard Steinbach
Abstract: We study video reconstruction from ultra-low-bitrate representations, where the primary challenge shifts from encoding to decoding. In this regime, reconstruction with classical and neural codecs introduces blur, while generative and semantic approaches often struggle to jointly preserve fidelity, temporal consistency, and perceptual quality. To address these limitations, we propose a causal video diffusion model that reconstructs videos from ultra-low-bitrate semantics and highly compressed frames by jointly modeling their complementary information. We further introduce temporal-only distillation from a bidirectional teacher to enable parameter-efficient training and causal few-step inference. Through extensive quantitative, qualitative, and subjective evaluation, we show that the proposed method outperforms classical, neural, generative, and semantic baselines in ultra-low-bitrate video reconstruction.
Authors: Tianyu Wang, Zhiyuan Ma, Qian Wang, Xinyi Zhang, Xinwei Long, Bowen Zhou
Abstract: Recent advancements in image generation have achieved impressive results in producing high-quality images. However, existing image generation models still generally struggle with a spatial reasoning dilemma, lacking the ability to accurately capture fine-grained spatial relationships from the prompt and correctly generate scenes with structural integrity. To mitigate this dilemma, we propose RL-RIG, a Reinforcement Learning framework for Reflection-based Image Generation. Our architecture comprises four primary components: Diffuser, Checker, Actor, and Inverse Diffuser, following a Generate-Reflect-Edit paradigm to spark the Chain of Thought reasoning ability in image generation for addressing the dilemma. To equip the model with better intuition over generation trajectories, we further develop Reflection-GRPO to train the VLM Actor for edit prompts and the Image Editor for better image quality under a given prompt, respectively. Unlike traditional approaches that solely produce visually stunning yet structurally unreasonable content, our evaluation metrics prioritize spatial accuracy, utilizing Scene Graph IoU and employing a VLM-as-a-Judge strategy to assess the spatial consistency of generated images on LAION-SG dataset. Experimental results show that RL-RIG outperforms existing state-of-the-art open-source models by up to 11% in terms of controllable and precise spatial reasoning in image generation.
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 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: 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: Wonsuk Jang, Thierry Tambe
Abstract: Diffusion Transformers (DiTs) achieve state-of-the-art video generation quality, but their substantial memory and computational footprints hinder edge deployment. Quantization can reduce these costs, yet existing methods often degrade video quality due to high activation variation and the difficulty of preserving semantic and temporal coherence. We propose SemanticDialect, which advances block-wise mixed-format quantization. In this framework, each block selects an optimal format (dialect) from a candidate set (formatbook), which is augmented with lookup tables that store quantization errors and quantized indices, enabling efficient per-block format selection and quantization with minimal online overhead. We further introduce attention-guided activation decomposition, which reduces quantization error via residual quantization, and semantic-aware dialect assignment (SeDA), which reduces cross-token quantization inconsistency by enforcing format uniformity among semantically correlated tokens. Experiments demonstrate that SemanticDialect outperforms prior quantization methods and block-wise formats (MXFP4, NVFP4) while approaching FP16 quality on Open-Sora 2.0. We also validate hardware deployability through RTL design and GPU kernel implementation.
Authors: Chenjun Li
Abstract: Multi-image reasoning remains a significant challenge for vision-language models (VLMs). We investigate a previously overlooked phenomenon: during chain-of-thought (CoT) generation, the text-to-image (T2I) attention of reasoning VLMs exhibits diffuse "pulses": sporadic and unfocused attention patterns that fail to concentrate on task-relevant images. We further reveal a systematic positional bias in attention allocation across images. Motivated by these observations, we propose PulseFocus, a training-free, inference-time method that structures CoT reasoning into interleaved plan/focus blocks with soft attention gating. By forcing the model to explicitly plan which image to examine and then gating decode-time attention to the referenced image, PulseFocus sharpens attention focus and yields consistent improvements on multi-image benchmarks like BLINK benchmark (+3.7%) and MuirBench (+1.07%).
Authors: Ziyu Chen, Fan Zhu, Hui Zhu, Deyi Kong, Xinkai Kuang, Yujia Zhang, Chunmao Jiang
Abstract: Recent 3D Gaussian Splatting (3DGS) methods have demonstrated the feasibility of self-driving scene reconstruction and novel view synthesis. However, most existing methods either rely solely on cameras or use LiDAR only for Gaussian initialization or depth supervision, while the rich scene information contained in point clouds, such as reflectance, and the complementarity between LiDAR and RGB have not been fully exploited, leading to degradation in challenging self-driving scenes, such as those with high ego-motion and complex lighting. To address these issues, we propose a robust and efficient LiDAR-reflectance-guided Salient Gaussian Splatting method (LR-SGS) for self-driving scenes, which introduces a structure-aware Salient Gaussian representation, initialized from geometric and reflectance feature points extracted from LiDAR and refined through a salient transform and improved density control to capture edge and planar structures. Furthermore, we calibrate LiDAR intensity into reflectance and attach it to each Gaussian as a lighting-invariant material channel, jointly aligned with RGB to enforce boundary consistency. Extensive experiments on the Waymo Open Dataset demonstrate that LR-SGS achieves superior reconstruction performance with fewer Gaussians and shorter training time. In particular, on Complex Lighting scenes, our method surpasses OmniRe by 1.18 dB PSNR.
Authors: Advaith Ravishankar, Serena Liu, Mingyang Wang, Todd Zhou, Jeffrey Zhou, Arnav Sharma, Ziling Hu, L\'eopold Das, Abdulaziz Sobirov, Faizaan Siddique, Freddy Yu, Seungjoo Baek, Yan Luo, Mengyu Wang
Abstract: State-of-the-art text-to-image models produce high-quality images, but inference remains expensive as generation requires several sequential ODE or denoising steps. Native one-step models aim to reduce this cost by mapping noise to an image in a single step, yet fair comparisons to multi-step systems are difficult because studies use mismatched sampling steps and different classifier-free guidance (CFG) settings, where CFG can shift FID, Inception Score, and CLIP-based alignment in opposing directions. It is also unclear how well one-step models scale to multi-step inference, and there is limited standardized out-of-distribution evaluation for label-ID-conditioned generators beyond ImageNet. To address this, we benchmark eight models spanning one-step flows (MeanFlow, Improved MeanFlow, SoFlow), multi-step baselines (RAE, Scale-RAE), and established systems (SiT, Stable Diffusion 3.5, FLUX.1) under a class-conditional protocol on ImageNet validation, ImageNetV2, and reLAIONet, our new proofread out-of-distribution dataset aligned to ImageNet label IDs. Using FID, Inception Score, CLIP Score, and Pick Score, we show that FID-focused model development and CFG selection can be misleading in few-step regimes, where guidance changes can improve FID while degrading text-image alignment and human preference signals, worsening visual quality. To make these tradeoffs explicit, we introduce CLIP-scaled and PickScore-scaled variants of FID (csFID, psFID) and Inception Score (csIS, psIS) to serve as a diagnostic for semantically aligned image generation.
Authors: Amy Rafferty, Rishi Ramaesh, Ajitha Rajan
Abstract: Deep learning models for chest X-ray diagnosis are constrained by limited coverage of clinically meaningful concept combinations in publicly available training datasets. While synthetic image generation has been explored to increase data diversity, existing methods rarely enforce clinical or anatomical constraints, limiting utility for improving model reliability. We propose CARPA, a clinically aware and anatomically grounded framework for synthetic chest X-ray generation that applies targeted perturbations to clinical concept vectors while preserving anatomical structure. By producing anatomically faithful synthetic images with controlled concept insertions and deletions, CARPA expands clinically relevant concept coverage. We evaluate CARPA across seven backbone architectures by fine-tuning models on synthetic subsets and testing on a held-out MIMIC-CXR benchmark. Compared to prior concept perturbation approaches, fine-tuning on CARPA-generated images consistently improves precision-recall performance, reduces predictive uncertainty, and improves model calibration. Structural and semantic analyses demonstrate high anatomical fidelity, strong concept alignment, and low semantic uncertainty. Evaluation by two expert radiologists further confirms realism and clinical agreement. Together, these results show that anatomically grounded concept perturbations enable more effective use of synthetic data, improving both performance and reliability of chest X-ray classification models and supporting safer clinical deployment.
Authors: Kaito Baba, Risa Kishikawa, Satoshi Kodera
Abstract: We propose MARL-Rad, a multi-modal multi-agent reinforcement learning framework for radiology report generation that trains the entire agentic system on policy within its deployed radiology workflow. MARL-Rad addresses the limitation of post-hoc agentization, where fixed LLMs are organized into hand-designed agentic workflows without being optimized for their assigned roles. Our framework decomposes chest X-ray interpretation into region-specific agents and a global integrating agent, and jointly optimizes them using clinically verifiable rewards. Experiments on the MIMIC-CXR and IU X-ray datasets show that MARL-Rad consistently improves clinical efficacy metrics such as RadGraph, CheXbert, and GREEN scores, achieving state-of-the-art clinical efficacy performance. Further analyses show that MARL-Rad improves laterality consistency and produces more accurate and detailed reports. A blinded clinician evaluation further suggests that MARL-Rad produces reports clinically comparable to ground-truth reports.
Authors: Jiayi Luo, Jiayu Chen, Jiankun Wang, Cong Wang, Hanxin Zhu, Qingyun Sun, Chen Gao, Zhibo Chen, Jianxin Li
Abstract: Diffusion Transformers (DiTs) achieve strong video generation quality but suffer from high inference cost due to dense 3D attention, motivating sparse attention techniques for improving efficiency. However, existing training-free sparse attention methods for video generation still face two unresolved limitations: ignoring layer heterogeneity in attention pruning and ignoring query-key coupling in block partitioning, which hinder a better quality-speedup trade-off. In this work, we uncover a critical insight: attention sparsity is an intrinsic layer-wise property, with only minor variation across different inputs. Motivated by this observation, we propose SVOO, a training-free sparse attention framework for fast video generation via offline layer-wise sparsity profiling and online bidirectional co-clustering. Specifically, SVOO adopts a two-stage paradigm: (i) offline layer-wise sensitivity profiling to derive intrinsic per-layer pruning levels, and (ii) online block-wise sparse attention via a bidirectional co-clustering algorithm. Extensive experiments on seven widely used video generation models demonstrate that SVOO achieves a superior quality-speedup trade-off over state-of-the-art methods, delivering up to 1.93x speedup while maintaining a PSNR of up to 29 dB on Wan2.1. Code is available at: https://github.com/Mutual-Luo/SVOO.
Authors: Bishoy Galoaa, Shayda Moezzi, Xiangyu Bai, Sarah Ostadabbas
Abstract: Recent video reasoning models increasingly produce spatio-temporal evidence chains that localize objects at specific timestamps. While these traces improve interpretability by grounding \emph{where} and \emph{when} evidence appears, they often leave the motion connecting observations, the \textit{how}, implicit. This makes dynamic and trajectory-dependent claims difficult to supervise, verify, or penalize when unsupported by the video. We formalize this missing component as Spatial-Temporal-Trajectory (STT) reasoning and introduce \textbf{Motion-o}, a motion-centric extension to vision-language models (VLMs) that makes trajectories explicit and verifiable. Motion-o augments evidence chains with Motion Chain of Thought (MCoT), a structured pathway that represents object motion through a discrete \texttt{
Authors: Shuxian Zhao, Jie Gui, Baosheng Yu, Dacheng Tao
Abstract: Steel surface defect analysis is critical for industrial quality control, yet existing benchmarks rely primarily on label-only annotations, limiting fine-grained semantic understanding and systematic evaluation of vision-language models. To address this gap, we introduce SteelDefectX, a vision-language dataset with multi-form textual annotations for steel surface defect analysis, comprising 7,778 images across 25 defect categories. At the class level, the dataset provides defect names, representative visual attributes, and industrial causes. At the sample level, each image is annotated with three forms of textual representations: (1) free-form natural language descriptions, (2) structured attribute annotations, and (3) template-based sentences. These annotations provide flexible textual supervision with varying levels of expressiveness and controllability. We further establish a comprehensive benchmark covering vision-language classification, segmentation, and cross-dataset transfer, along with additional evaluations such as retrieval and text-guided localization. Experimental results reveal a trade-off between structure and flexibility in textual representations. Structured attributes provide more stable semantic alignment, while natural language descriptions improve transferability and fine-grained spatial grounding. These findings highlight the critical role of textual design in industrial vision-language learning. SteelDefectX provides a new benchmark for studying semantic alignment and generalization in industrial vision-language learning. The code and dataset are available at https://github.com/Zhaosxian/SteelDefectX.
Authors: Vardaan Pahuja, Samuel Stevens, Alyson East, Sydne Record, Yu Su
Abstract: Morphological traits are physical characteristics of biological organisms that provide vital clues on how organisms interact with their environment. Yet extracting these traits remains a slow, expert-driven process, limiting their use in large-scale ecological studies. A major bottleneck is the absence of high-quality datasets linking biological images to trait-level annotations. In this work, we demonstrate that sparse autoencoders trained on foundation-model features yield monosemantic, spatially grounded neurons that consistently activate on meaningful morphological parts. Leveraging this property, we introduce a trait annotation pipeline that localizes salient regions and uses vision-language prompting to generate interpretable trait descriptions. Using this approach, we construct Bioscan-Traits, a dataset of 80K trait annotations spanning 19K insect images from BIOSCAN-5M. Human evaluation confirms the biological plausibility of the generated morphological descriptions. We assess design sensitivity through a comprehensive ablation study, systematically varying key design choices and measuring their impact on the quality of the resulting trait descriptions. By annotating traits with a modular pipeline rather than prohibitively expensive manual efforts, we offer a scalable way to inject biologically meaningful supervision into foundation models, enable large-scale morphological analyses, and bridge the gap between ecological relevance and machine-learning practicality.
Authors: Siheng Wang, Yanshu Li, Bohan Hu, Zhengdao Li, Haibo Zhan, Linshan Li, Weiming Liu, Ruizhi Qian, Guangxin Wu, Hao Zhang, Jifeng Shen, Piotr Koniusz, Zhengtao Yao, Junhao Dong, Qiang Sun
Abstract: Open-vocabulary Object Detection (OVOD) enables models to recognize objects beyond predefined categories, but existing approaches remain limited in practical deployment. On the one hand, multimodal designs often incur substantial computational overhead due to their reliance on text encoders at inference time. On the other hand, tightly coupled training objectives introduce a trade-off between closed-set detection accuracy and open-world generalization. Thus, we propose Decoupled Cognition DETR (DeCo-DETR), a vision-centric framework that addresses these challenges through a unified decoupling paradigm. Instead of depending on online text encoding, DeCo-DETR constructs a hierarchical semantic prototype space from region-level descriptions generated by pre-trained LVLMs and aligned via CLIP, enabling efficient and reusable semantic representation. Building upon this representation, the framework further disentangles semantic reasoning from localization through a decoupled training strategy, which separates alignment and detection into parallel optimization streams. Extensive experiments on standard OVOD benchmarks demonstrate that DeCo-DETR achieves competitive zero-shot detection performance while significantly improving inference efficiency. These results highlight the effectiveness of decoupling semantic cognition from detection, offering a practical direction for scalable OVOD systems.
Authors: Daniele Solombrino, Antonio Andrea Gargiulo, Alessandro Zirilli, Luca Zhou, Adrian Robert Minut, Emanuele Rodol\`a
Abstract: We show that robustness to post-training quantization (PTQ) is a transferable direction in weight space. We call this direction the quantization vector: extracted from a donor task by simple weight-space arithmetic, it can be used to patch a receiver model and improve post-PTQ Top-1 accuracy by up to 60 points in a 3-bit setting, without receiver-side quantization-aware training (QAT). Because the method requires no receiver training data, it provides a zero-shot, low-cost alternative to QAT for extremely low-bit deployment. Across four ViT scales and 22 image classification tasks, donor quantization vectors often yield substantial gains even when donor and receiver tasks differ markedly. We further prove rigorously that quantization vectors are well-defined and do not suffer from reparameterization symmetries, and provide a local geometric account of their effect. Together, these results suggest that quantization robustness can be partially isolated, reused, and transferred through simple weight-space algebra.
Authors: Jing Jin, Hao Liu, Yan Bai, Yihang Lou, Zhenke Wang, Tianrun Yuan, Juntong Chen, Yongkang Zhu, Fanhu Zeng, Xuanyu Zhu, Tao Feng, Yige Xu
Abstract: Multimodal large language models (MLLMs) have shown promising reasoning abilities, yet evaluating their performance in specialized domains remains challenging. STEM reasoning is a particularly valuable testbed because it provides highly verifiable feedback, but existing benchmarks often permit unimodal shortcuts due to modality redundancy and focus mainly on final-answer accuracy, overlooking the reasoning process itself. To address this challenge, we introduce StepSTEM: a graduate-level benchmark of 283 problems across mathematics, physics, chemistry, biology, and engineering for fine-grained evaluation of cross-modal reasoning in MLLMs. StepSTEM is constructed through a rigorous curation pipeline that enforces strict complementarity between textual and visual inputs. We further propose a general step-level evaluation framework for both text-only chain-of-thought and interleaved image-text reasoning, using dynamic programming to align predicted reasoning steps with multiple reference solutions. Experiments across a wide range of models show that current MLLMs still rely heavily on textual reasoning, with even Gemini 3.1 Pro and Claude Opus 4.6 achieving only 38.29% accuracy. These results highlight substantial headroom for genuine cross-modal STEM reasoning and position StepSTEM as a benchmark for fine-grained evaluation of multimodal reasoning. Source code is available at https://github.com/lll-hhh/STEPSTEM.
Authors: Shyang-En Weng, Yi-Cheng Liao, Yu-Syuan Xu, Wei-Chen Chiu, Ching-Chun Huang
Abstract: Pretrained diffusion models have revolutionized real-world image super-resolution (Real-ISR) but suffer from computational bottlenecks due to iterative sampling. Recent single-step distillation accelerates inference but faces a stark perception-distortion trade-off due to rigid timestep initialization, distributional trajectory mismatches, and fragile stochastic modulation. To address this, we present Adaptive Inversion and Degradation-aware Sampling for Real-ISR (IDaS-SR), a one-step framework bridging the deterministic restoration and stochastic generation manifolds. At its core, the Manifold Inversion Noise Estimator (MINE) resolves these initialization and trajectory mismatches by predicting a severity-aware timestep and inversion noise, precisely anchoring low-quality latents onto the diffusion trajectory. Furthermore, to mitigate fragile stochastic modulation, we propose CHARIOT, a continuous generative steering mechanism. By rescheduling trajectories and interpolating noise, it enables explicit navigation of the perception-distortion boundary without compromising structural priors. Extensive experiments demonstrate that IDaS-SR outperforms state-of-the-art methods, seamlessly transitioning from a rigorous structural restorer to a sophisticated texture hallucinator in a single inference step.
Authors: Enqi Liu, Liyuan Pan, Zhi Gao, Yan Yang, Chenrui Shi, Yang Liu, Jingrong Wu, Qing Li
Abstract: Recent advancements in Graphical User Interface (GUI) agents have predominantly focused on training paradigms like supervised fine-tuning (SFT) and reinforcement learning (RL). However, the challenge of high-dynamic GUI environments remains largely underexplored. Existing agents typically rely on a single screenshot after each action for decision-making, leading to a partially observable (or even unobservable) Markov decision process, where the key GUI state including important information for actions is often inadequately captured. To systematically explore this challenge, we introduce DynamicGUIBench, a comprehensive online GUI benchmark spanning ten applications and diverse interaction scenarios characterized by important interface changes between actions. Furthermore, we present DynamicUI, an agent designed for dynamic interfaces, which takes screen-recording videos of the interaction process as input and consists of three components: a dynamic perceiver, a refinement strategy, and a reflection. Specifically, the dynamic perceiver clusters frames of the GUI video, generates captions for the centroids, and iteratively selects the most informative frames as the salient dynamic context. Considering that there may be inconsistencies and noise between the selected frames and the textual context of the agent, the refinement strategy employs an action-conditioned filtering to refine thoughts to mitigate thought-action inconsistency and redundancy. Based on the refined agent trajectories, the reflection module provides effective and accurate guidance for further actions. Experiments on DynamicGUIBench demonstrate that DynamicUI significantly improves the performance in dynamic GUI environments, while maintaining competitive performance on other public benchmarks.
Authors: Yashwant Pravinrao Bangde, Debaditya Roy
Abstract: Vision-Language Models (VLMs) exhibit strong performance in instruction following and open-ended vision-language reasoning, yet they frequently generate fluent outputs that are weakly grounded in visual evidence. Prior works have shown that instruction prompting further worsens this issue by amplifying language priors, especially when the visual signal is uncertain or ambiguous. To address this challenge, we propose a decoding framework that explicitly balances linguistic informativeness and visual faithfulness during generation. Our method, Instruction-Evidence Contrastive Dual-Stream Decoding (IECD$^2$), maintains two parallel probability distribution of tokens at each decoding step: an instruction-driven stream that promotes expressive and informative responses, and an evidence-driven stream that enforces strict grounding in the image. These two streams are adaptively fused using a symmetric KL-based contrastive gate, which suppresses tokens favored by language priors but unsupported by visual evidence, while preserving them when both distributions agree. We evaluate IECD$^2$ on multiple datasets spanning various generative vision-language reasoning tasks such as captioning and visual question answering on multiple datasets such as, POPE, MME, VQAv2, AMBER, and MSCOCO. IECD$^2$ demonstrates consistent improvements in task accuracy and reasoning performance with substantial reduction in hallucination compared to state-of-the-art decoding approaches.
Authors: Siyuan Huang, Xiaoye Qu, Yafu Li, Tong Zhu, Zefeng He, Muxin Fu, Daizong Liu, Wei-Long Zheng, Yu Cheng
Abstract: While autoregressive Large Vision-Language Models (LVLMs) demonstrate remarkable proficiency in multimodal tasks, they face a "Visual Signal Dilution" phenomenon, where the accumulation of textual history expands the attention partition function, causing visual attention to decay inversely with generated sequence length. To counteract this, we propose Persistent Visual Memory (PVM), a lightweight learnable module designed to strengthen sustained, on-demand access to visual evidence. Integrated as a parallel branch alongside the Feed-Forward Network (FFN) in LVLMs, PVM establishes a distance-agnostic retrieval pathway that directly provides visual embeddings for enhanced visual perception, thereby structurally mitigating the signal suppression inherent to deep generation. Extensive experiments on Qwen3-VL models demonstrate that PVM brings notable improvements with negligible parameter overhead, delivering consistent average accuracy gains across both 4B and 8B scales, particularly in complex reasoning tasks that demand persistent visual perception. Furthermore, in-depth analysis reveals that PVM shows improved robustness in longer generations and accelerates internal prediction convergence.
Authors: Roberto Isai Navaro-Avi\~na, Eduardo Said Merin-Martinez, Andres Mendez-Vazquez, Eduardo Rodriguez-Tello
Abstract: Single-Image Super-Resolution (SISR) aims to reconstruct a High-Resolution (HR) image from a Low-Resolution (LR) observation, a fundamentally ill-posed problem where high-frequency details are severely degraded at large upscaling factors. Recent advances have been driven by transformer-based architectures and diffusion models improve global context modeling and perceptual quality at the cost of increased computational complexity. In contrast, this work focuses on enhancing the expressivity of local operators under minimal resources. We propose SRGAN--CKAN, a hybrid super-resolution framework that integrates Convolutional Kolmogorov--Arnold Networks (CKAN) into an adversarial learning setting reformulating convolution as a nonlinear patch-based transformation. The proposed operator replaces linear local mappings with spline-based functional representations, allowing expressive modeling of complex local structures and high-frequency textures using minimal hardware resources. Experimental results demonstrate that the proposed approach improves perceptual quality while preserving reconstruction fidelity, achieving a favorable balance between distortion-based and perceptual metrics. These results are obtained under constrained computational settings, highlighting the efficiency of the proposed formulation. Overall, this work introduces a complementary direction to existing approaches by improving the representational power of local transformations, providing an efficient and scalable alternative to globally intensive architectures.
Authors: Jinyuan Shao, Sangyoong Park, Chunxi Zhao, Ayman Habib, Songlin Fei
Abstract: Airborne Laser Scanning (ALS) can collect point clouds across large areas, enabling large-scale forest inventory. However, ALS point clouds are sparse and noisy, resulting in inaccurate individual-tree-level forest inventory, such as stem localization and tree size estimation. To overcome this problem, we propose a deep learning model, 3D Forest Super Resolution (3DFSR), to simultaneously improve point density and reduce noise for ALS forest point cloud. 3DFSR is a voxel-based CNN with a U-Net architecture. The proposed 3DFSR is evaluated on ALS point clouds collected in both temperate forests in the U.S. and boreal forests in Germany. Experimental results demonstrate that 3DFSR can generate finer point clouds of tree structure than other state-of-the-art point cloud super-resolution algorithms, achieving 0.249 m Chamfer Distance and 2.711 m Hausdorff Distance. Furthermore, to verify the effectiveness of 3DFSR point clouds in forest inventory, we conduct stem detection, DBH measurements, and stem reconstruction on both original ALS point clouds and 3DFSR enhanced point clouds. We find that stem detection and reconstruction algorithms developed for TLS/MLS point clouds can directly work on our 3DFSR point clouds, and DBH can be derived with circle-fitting method. F1 score of stem detection is improved from 0.71 on original ALS point clouds to 0.97 on 3DFSR point clouds; DBH estimation improves from 13.45 cm RMSE using allometric equations to 6.43 cm using circle fitting; comparing to stems reconstruction from MLS point clouds, stem reconstructed from 3DFSR point clouds has 0.170 m of Chamfer Distance and 0.377 m of Hausdorff Distance, and 0.95 R2 volume estimation. Finally, we find that the proposed 3DFSR is applicable to process point densities from 10 to 1700 points/m2; it also can be generalized across data collected from different LiDAR platforms without transfer learning.
Authors: Abdullah Ahmad Khan, Hamid Laga, Ferdous Sohel
Abstract: Machine unlearning in Vision-Language Models (VLMs) is required for compliance with the General Data Protection Regulation (GDPR), yet current evaluation practices are inconsistent. We present the first systematic study of metric reliability in multimodal unlearning. Five standard metrics, Forget Accuracy (FA), Retain Accuracy (RA), Membership Inference Attack (MIA), Activation Distance (AD), and JS divergence (JS), yield conflicting method rankings across three VQA benchmarks (MLLMU-Bench, UnLOK-VQA, MMUBench). Kendall tau analysis over 36 unlearned LLaVA-1.5-7B models reveals two opposing clusters, {FA, RA, MIA} and {AD, JS}, with tau_FA_AD = -0.26, reproduced on BLIP-2 OPT-2.7B. Agreement is lower in multimodal VQA (average tau = 0.086) than in unimodal classification (average tau = 0.158; difference = 0.072), indicating that dual image-and-text pathways amplify inconsistency. We introduce the Unified Quality Score (UQS), a composite metric with weights derived from each metric's Spearman correlation with the oracle distance d(M_hat, M_star), where M_star is the oracle model retrained only on the retain set. RA shows the strongest reliability (rho = 0.484, p = 0.003), while FA is negatively correlated (rho = -0.418, p = 0.011). UQS yields stable rankings under 100 random weight perturbations (tau = 0.647 +- 0.262). We release the benchmark, 36 checkpoints, and an interactive leaderboard. Code and pre-computed results are available at https://github.com/neurips26/UnifiedUnl.
Authors: Jiaqi Shi, Jin Xiao, Xiaoguang Hu, Wenxuan Ji, Zichong Jia, Zifan Long, Tianyou Chen, Baochang Zhang
Abstract: In 3D point cloud understanding, the core challenge lies in accurately capturing discriminative features within complex neighborhoods, which directly affects the execution precision of downstream tasks such as embodied AI and autonomous driving. Existing methods explore feature correlation discrimination but are limited to point-level spatial distribution or channel responses, enabling only coarse-grained level evaluation. For modern multi-scale point cloud networks, such coarse-grained metrics inevitably incur significant information loss in deeper layers. To address this, we propose PointCRA, a novel network with a channel-level metric-based enhancement mechanism. Our core idea is to introduce temporal trend variation as a new evaluation dimension to avoid the information loss caused by weight dimension collapse in existing spatial and channel attention mechanisms. On this basis, we construct a multi-level calibration framework guided by neighborhood homogeneity for weight calibration, and design a dedicated loss function to enhance channel discriminability.PointCRA leverages intrinsic feature priors to adaptively correct feature aggregation, offering interpretability with low parameter overhead. Our method is transferable, interpretable, and efficient. We validate the proposed method on diverse datasets and benchmark models, and further demonstrate its rationality through extensive analytical experiments. Our PointCRA achieves 77.5\% mIoU on the S3DIS dataset, 90.4\% OA on the ScanObjectNN dataset, and 87.4\% instance mIoU on the ShapeNetPart dataset. The code and pretrained weights are publicly available on GitHub: https://github.com/AGENT9717/PointCRA
Authors: Evangelos Ntavelis, Sean Wu, Mohamad Shahbazi, Fabio Maninchedda, Dmitry Kostiaev, Artem Sevastopolsky, Vittorio Megaro, Trevor Phillips, Alejandro Blumentals, Shridhar Ravikumar, Mehak Gupta, Reinhard Knothe, Jeronimo Bayer, Matthias Vestner, Simon Schaefer, Thomas Etterlin, Christian Zimmermann, Mathias Deschler, Peter Kaufmann, Stefan Brugger, Sebastian Martin, Brian Amberg, Tom Runia
Abstract: We propose HeadsUp, a scalable feed-forward method for reconstructing high-quality 3D Gaussian heads from large-scale multi-camera setups. Our method employs an efficient encoder-decoder architecture that compresses input views into a compact latent representation. This latent representation is then decoded into a set of UV-parameterized 3D Gaussians anchored to a neutral head template. This UV representation decouples the number of 3D Gaussians from the number and resolution of input images, enabling training with many high-resolution input views. We train and evaluate our model on an internal dataset with more than 10,000 subjects, which is an order of magnitude larger than existing multi-view human head datasets. HeadsUp achieves state-of-the-art reconstruction quality and generalizes to novel identities without test-time optimization. We extensively analyze the scaling behavior of our model across identities, views, and model capacity, revealing practical insights for quality-compute trade-offs. Finally, we highlight the strength of our latent space by showcasing two downstream applications: generating novel 3D identities and animating the 3D heads with expression blendshapes.
Authors: James Walsh, William Fawcett, Grace Colvard, Ra\'ul Ramos-Poll\'an
Abstract: Semantic search of Earth observation archives remains challenging. Visual foundation models such as CLAY produce rich embeddings of satellite imagery but lack the natural-language grounding needed for intuitive query, and full contrastive training of a remote-sensing CLIP-style model requires paired data and compute that are unavailable at global scale. To allow natural language querying at global scales, we present GeoQuery, a zero-shot retrieval system that sidesteps data and compute constraints through a two-stage semantic and visual search, leveraging a natural language embedding of a subset (proxy) of global data. Rather than training a joint encoder, we generate language descriptions for a 100k proxy subset of global Sentinel-2 tiles and optimise the description-generation prompt so that distances in the resulting text-embedding space correlate with distances in the frozen CLAY visual-embedding space. Queries are resolved in two stages, with a text-similarity search over the proxy subset followed by a visual nearest-neighbour search over worldwide CLAY embeddings On 76 disaster-location queries covering UK floods, US wildfires, and US droughts, GeoQuery achieves 31.6\% accuracy within 50\,km, with the strongest performance on floods (50\% within 50\,km) where terrain features are well captured by RGB embeddings. Deployed within a crisis response system called \ECHO{}, GeoQuery identified vulnerable areas during Brisbane's 2025 Cyclone Alfred, with downstream flood simulations reproducing historical patterns. Prompt-aligned proxies offer a practical bridge between EO foundation models and operational retrieval when full contrastive training is out of reach.
Authors: Dongfang Zhao
Abstract: Vector search systems built on frozen vision encoders face queries from unseen classes at deployment, yet existing adapter training collapses under this shift: high-capacity adapters with global contrastive losses silently reassign unseen-class samples to wrong seen-class clusters, dropping worst-case Label Precision by over 40 points below the frozen baseline in our tests. We propose Euclidean Geodesic Alignment (EGA), a residual adapter that couples three principles: zero initialization, local triplet loss, and hypersphere projection. These collectively induce a self-limiting dynamic: triplets that already satisfy a small margin stop producing gradients, so the adapter automatically stops updating where the local geometry is already correct. Our experiments show that at convergence $96.5\%$ of triplets are gradient-free, leaving unseen-class regions largely untouched while still enabling full-capacity refinement of seen classes. Across five diverse out-of-distribution (OOD) benchmarks, EGA achieves the highest worst-case Label Precision on the four primary splits and a consistent improvement on the fifth. The design also transfers to stronger backbones in addition to CLIP, and we provide an analytical justification linking gradient sparsity to bounded OOD perturbation.
Authors: Kuanwei Lin, Wenhao Zhang, Ge Li
Abstract: Video large multimodal models increasingly face a scalability bottleneck: long videos produce excessively long visual-token sequences, which sharply increase memory and latency during inference. While existing compression methods are effective in specific settings, most are either weakly query-aware or apply a fixed compression policy across frames, proving suboptimal when visual evidence is unevenly distributed over time. To address this, we present VideoRouter, a query-adaptive dual-router framework built on InternVL for budgeted evidence allocation. The Semantic Router predicts the dominant allocation policy, choosing between broad temporal coverage and adaptive high-resolution preservation, while the Image Router uses early LLM layers to score frame relevance. This enables aggressive compression on less relevant frames while preserving detail on critical evidence frames. To train both routers, we build Video-QTR-10K for allocation-policy supervision and Video-FLR-200K for frame-relevance supervision. Experiments on VideoMME, MLVU, and LongVideoBench show that VideoRouter consistently improves over the InternVL baseline under comparable or lower budgets, achieving up to a 67.9% token reduction.
Authors: Hao Lin, Kunyang Lv, Xu Jiang, Jingqi Tian, Zhongjing Du, Jiayu Ding, Qiaoman Zhang, Hongbo Jin
Abstract: Training VideoLLMs for complex reasoning remains challenging due to sparse sequence level rewards and the lack of fine grained credit assignment over long, temporally grounded reasoning trajectories. While reinforcement learning with verifiable rewards (RLVR) provides reliable supervision, it fails to capture token level contributions, leading to inefficient learning. Conversely, existing self distillation methods offer dense supervision but lack structure and diagnostic specificity, and often interact unstably with reinforcement learning. In this work, we propose VISD, a structured self distillation framework that introduces diagnostically meaningful privileged information for video reasoning. VISD employs a video aware judge model to decompose reasoning quality into multiple dimensions, including answer correctness, logical consistency, and spatio-temporal grounding, and uses this structured feedback to guide a teacher policy for token level supervision. To stably integrate dense supervision with RL, we introduce a direction magnitude decoupling mechanism, where rollout level advantages computed from rewards determine update direction, while structured privileged signals modulate token level update magnitudes. This design enables semantically aligned and fine grained credit assignment, improving both reasoning faithfulness and training efficiency. Additionally, VISD incorporates curriculum scheduling and EMA based teacher stabilization to support robust optimization over long video sequences. Experiments on diverse benchmarks show that VISD consistently outperforms strong baselines, improving answer accuracy and spatio temporal grounding quality. Notably, VISD reaches these gains with nearly 2x faster convergence in optimization steps, highlighting the effectiveness of structured self supervision in improving both performance and sample efficiency for VideoLLMs.
Authors: Abdelrahman Zaian, Sheethal Bhat, Mohamed Abdalkader, Andreas Maier
Abstract: Diabetic Retinopathy (DR) is a leading cause of preventable blindness among working-age adults worldwide, yet most automated screening systems are limited to image-level classification and lack clinically structured reporting. We propose Retina-RAG, a low-cost modular framework that jointly performs DR severity grading, macular edema (ME) detection, and report generation. The architecture decouples a high-performance retinal classifier and a parameter-efficient vision-language model (Qwen2.5-VL-7B-Instruct) adapted via Low-Rank Adaptation (LoRA), enabling flexible component integration. A retrieval-augmented generation (RAG) module injects curated ophthalmic knowledge together with structured classifier outputs at inference time to improve diagnostic consistency and reduce hallucinations. Retina-RAG achieves an F1-score of 0.731 for DR grading and 0.948 for ME detection, substantially outperforming zero-shot Qwen (0.096, 0.732) and MMed-RAG (0.541, 0.641) on a retinal disease detection dataset with captions. For report generation, Retina-RAG attains ROUGE-L 0.438 and SBERT similarity 0.884, exceeding all baselines. The full framework operates on a single consumer-grade GPU, demonstrating that clinically structured retinal AI can be achieved with modest computational resources.
Authors: Thong Nguyen, Khoi M. Le, Cong-Duy Nguyen, Luu Anh Tuan, See-Kiong Ng, Chunyan Miao
Abstract: Recent advancements in image animation have utilized diffusion models to breathe life into static images. However, existing controllable frameworks typically rely on Lagrangian motion guidance, where optical flow is estimated relative to the initial frame. This paper revisits the same optical-flow primitive through a more local supervision design: we use adjacent-frame Eulerian motion fields to guide generation, where the motion signal always describes a short temporal hop. This shift enables parallelized training and provides bounded-error supervision throughout the generation process. To mitigate the drift artifacts common in adjacent frame generation, we introduce a Bidirectional Geometric Consistency mechanism, which computes a forward-backward cycle check to mathematically identify and mask occluded regions, preventing the model from learning incorrect warping objectives. Extensive experiments demonstrate that our approach accelerates training, preserves temporal coherence, and reduces dynamic artifacts compared to reference-based baselines.
Authors: Roussel Desmond Nzoyem, Mauro Comi
Abstract: Training world models on vast quantities of unlabelled videos is a critical step toward fully autonomous intelligence. However, the prevailing paradigm of encoding raw pixels into opaque latent spaces and relying on heavy decoders for reconstruction leaves these models computationally expensive and uninterpretable. We address this problem by introducing NOVA, a world modelling framework that represents the system state as the weights and biases of an auxiliary coordinate-based implicit neural representation (INR). This structured representation is analytically rendered, which eliminates the decoder bottleneck while conferring compactness, portability, and zero-shot super-resolution. Furthermore, like most latent action models, NOVA can be distilled into a context-dependent video generator via an action-matching objective. Surprisingly, without resorting to auxiliary losses or adversarial objectives, NOVA can disentangle structural scene components such as background, foreground, and inter-frame motion, enabling users to edit either content or dynamics without compromising the other. We validate our framework on several challenging datasets, achieving strong controllable forecasting while operating on a single consumer GPU at $\sim$40M parameters. Ultimately, structured representations like INRs not only enhance our understanding of latent dynamics but also pave the way for immersive and customisable virtual experiences.
Authors: Dijia Zhan, Jinyi Li, Chenxi Zheng, Shaoyu Huang, Yong Li, Jie Tang, Xuemiao Xu
Abstract: Existing Vision-Language Navigation (VLN) methods typically adopt an egocentric, step-by-step paradigm, which struggles with error accumulation and limits efficiency. While recent approaches attempt to leverage pre-built environment maps, they often rely on incrementally updating memory graphs or scoring discrete path proposals, which restricts continuous spatial reasoning and creates discrete bottlenecks. We propose Top-Down VLN (TD-VLN), reformulating navigation as a one-step global path planning problem on pre-built top-down maps, supported by our newly constructed R2R-TopDown dataset. To solve this, we introduce NavOne, a unified framework that directly predicts dense path probabilities over multi-modal maps in a single end-to-end forward pass. NavOne features a Top-Down Map Fuser for joint multi-modal map representation, and extends Attention Residuals for spatial-aware depth mixing. Extensive experiments on R2R-TopDown show that NavOne achieves state-of-the-art performance among map-based VLN methods, with a planning-stage speedup of 8x over existing map-based baselines and 80x over egocentric methods, enabling highly efficient global navigation.
Authors: Afonso Oliveira, Jo\~ao P. Matos-Carvalho, Filipe Moutinho, Nuno Fachada
Abstract: The increasing frequency and severity of wildfires necessitates advanced methods for effective surveillance and management, as traditional ground-based techniques often struggle to adapt to rapidly changing fire behavior and environmental conditions. This study investigates the use of multispectral aerial and satellite imagery for wildfire management through an assessment of current literature and two practical case studies. We evaluate several multispectral indices for their ability to extract environmental features critical for analyzing wildfire behavior, including vegetation, water bodies, and artificial structures. Our results highlight NVDI for vegetation, MNDWI for water features, and MSR for artificial structures as particularly effective for segmentation and feature extraction. The application of these indices enhances wildfire data processing and supports improved monitoring, risk assessment, and response strategies, demonstrating the potential of multispectral imagery to complement traditional wildfire monitoring and management approaches.
Authors: Hongyuan Zhang, Yanchen Xu, Sida Huang, Xuelong Li
Abstract: Inspired by the idea of Positive-incentive Noise (Pi-Noise or $\pi$-Noise) that aims at learning the reliable noise beneficial to tasks, we scientifically investigate the connection between contrastive learning and $\pi$-noise in this paper. By converting the contrastive loss to an auxiliary Gaussian distribution to quantitatively measure the difficulty of the specific contrastive model under the information theory framework, we properly define the task entropy, the core concept of $\pi$-noise, of contrastive learning. It is further proved that the predefined data augmentation in the standard contrastive learning paradigm can be regarded as a kind of point estimation of $\pi$-noise. Inspired by the theoretical study, a framework that develops a $\pi$-noise generator to learn the beneficial noise (instead of estimation) as data augmentations for contrast is proposed. The designed framework can be applied to diverse types of data and is also completely compatible with the existing contrastive models. From the visualization, we surprisingly find that the proposed method successfully learns effective augmentations. Our code is available at https://github.com/hyzhang98/PiNDA.
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: Dikshant Sagar, Kaiwen Yu, Alejandro Yankelevich, Jianming Bian, Pierre Baldi
Abstract: Recent advances in Large Language Models (LLMs) have demonstrated their remarkable capacity to process and reason over structured and unstructured data modalities beyond natural language. In this work, we explore the applications of Vision Language Models (VLMs), specifically a fine-tuned variant of LLaMA 3.2 to the task of identifying neutrino interactions in pixelated detector data from high-energy physics (HEP) experiments. We benchmark this model against a state-of-the-art convolutional neural network (CNN) architecture, similar to those used in major neutrino experiments, which have achieved high efficiency and purity in classifying electron and muon neutrino events, and also a Vision Transformer (ViT-h/14), which is the same architecture inside the VLM's vision encoder. Our evaluation considers both classification performance and interpretability of the model predictions, comparing a VLM with a vision-only transformer (ViT) and a convolutional neural network (CNN) baseline. We find that transformer-based architectures outperform conventional CNNs in classification accuracy and robustness, with the VLM providing additional flexibility through the integration of auxiliary textual or semantic information and enabling more interpretable, reasoning-based predictions. These results highlight the potential of large transformer models, particularly vision-language models, as general-purpose backbones for physics event classification, combining strong performance, robustness, and interpretability, and opening new avenues for multimodal reasoning in experimental neutrino physics.
Authors: Max Hartman, Vidhata Jayaraman, Moulik Choraria, Akhil Bhimaraju, Lav R. Varshney
Abstract: Vision-language models achieve incredible performance across a wide range of tasks, but their large size makes inference costly. Recent work has shown that multimodal processing contains significant redundancies, making it possible to skip certain layers with minimal performance loss. Yet current pruning techniques remain ad-hoc, relying on heuristics or hyperparameter sweeps rather than principled criteria for determining when layer skipping is beneficial. In this paper, we propose a unified framework that characterizes the redundancy conditions under which pruning can enhance efficiency without sacrificing performance. Central to our approach are experimentally verifiable and interpretable notions of redundancy that can be evaluated without requiring downstream task performance as a metric. Applying this framework, we corroborate prior findings that both early and late vision tokens are redundant across models, and we validate our conditions by showing they align with actual performance degradation. Beyond these empirical results, our framework provides a theoretically grounded understanding of redundancy in VLMs and unifies many of the ideas behind modern layer-skipping techniques.
Authors: Ali Yavari, Alireza Mohamadi, Elham Beydaghi, Philipp Seeb\"ock, Rainer A. Leitgeb
Abstract: State-of-the-art attribution methods rely on adversarial sample generation that applies an all-pass filter across the frequency spectrum, discarding fine-grained high-frequency information that is demonstrably important for accurate feature attribution in deep neural networks. By generating adversarial samples that selectively perturb high- and low-frequency components, we can probe which spectral features a model relies on most -- directly translating frequency-domain exploration into attribution signals. Building on this insight, we propose FAMPE (Frequency-Aware Model Parameter Explorer), a novel attribution method that introduces an FFT-based \alpha-weighted perturbation scheme -- separately modulating high- and low-frequency components via an energy-driven spectral cutoff -- and, crucially, integrates this frequency-aware exploration directly into model parameter exploration for attribution, a connection that has not been established in prior work. Unlike prior frequency-aware adversarial approaches that target transferability or imperceptibility, FAMPE's specific formulation is designed and validated exclusively for explainability, translating spectral structure into fine-grained attribution maps without requiring any manual baseline selection. Evaluated on ImageNet across four architectures spanning CNNs and Vision Transformers, at fixed \alpha = 0.1 FAMPE outperforms AttEXplore by 4.25% on Inception-v3 and 12.04% on MaxViT-T, with per-sample oracle selection further revealing that low-frequency-dominated images systematically benefit from high-frequency perturbations -- underscoring the potential of adaptive spectral exploration. Our ablation studies confirm that high-frequency perturbations are disproportionately responsible for attribution precision, while excessive low-frequency noise degrades global structural coherence.
Authors: Yongxian Wei, Yilin Zhao, Zixuan Hu, Li Shen, Xinrui Chen, Runxi Cheng, Sinan Du, Hao Yu, Chun Yuan, Dian Li
Abstract: Data synthesis for training large reasoning models offers a scalable alternative to limited, human-curated datasets, enabling the creation of high-quality data. However, existing approaches face several challenges: (i) indiscriminate generation that ignores the solver's ability and yields low-value problems, or reliance on complex data pipelines to balance problem difficulty; and (ii) a lack of reasoning in problem generation, leading to shallow problem variants. In this paper, we develop a problem generator that reasons explicitly to plan problem directions before synthesis and adapts difficulty to the solver's ability. Specifically, we construct related problem pairs and augment them with intermediate problem-design CoT produced by a reasoning model. These data are used to bootstrap problem-design strategies in the generator. Then, we treat the solver's feedback on synthetic problems as a reward signal, enabling the generator to calibrate difficulty and produce complementary problems near the edge of the solver's competence. Extensive experiments on 10 mathematical and general reasoning benchmarks show that our proposed framework achieves a cumulative average improvement of 3.4%, demonstrating robust generalization across both language and vision-language models.
Authors: Yuxiang Wan, Ryan Devera, Wenjie Zhang, Ju Sun
Abstract: Foundation flow-matching (FM) models promise a universal prior for solving inverse problems (IPs), yet today they trail behind domain-specific or even untrained priors. How can we unlock their potential? We introduce FMPlug, a plug-in framework that redefines how foundation FMs are used in IPs. FMPlug combines an instance-guided, time-dependent warm-start strategy with a sharp Gaussianity regularization, adding problem-specific guidance while preserving the Gaussian structures. This leads to a significant performance boost across image restoration and scientific IPs. Our results point to a path for making foundation FM models practical, reusable priors for IP solving.
Authors: Boyuan Chen, Tianyuan Zhang, Haoran Geng, Caiyi Zhang, Peihao Li, Kiwhan Song, William T. Freeman, Jitendra Malik, Pieter Abbeel, Russ Tedrake, Vincent Sitzmann, Yilun Du
Abstract: General-purpose robots require decision-making models that generalize across diverse tasks and environments. Recent works build robot foundation models by extending multimodal large language models (MLLMs) with action outputs, creating vision-language-action (VLA) systems. These efforts are motivated by the intuition that MLLMs' large-scale language and image pretraining can be effectively transferred to the action output modality. In this work, we explore an alternative paradigm of using large-scale video pretraining as a primary modality for building robot foundation models. Unlike static images and language, videos capture spatio-temporal sequences of states and actions in the physical world that are naturally aligned with robotic behavior. We curate an internet-scale video dataset of human activities and task demonstrations, and train, for the first time at a foundation-model scale, an open video model for generative robotics planning. The model produces zero-shot video plans for novel scenes and tasks, which we post-process to extract executable robot actions. We evaluate task-level generalization through third-party selected tasks in the wild and real-robot experiments, demonstrating successful physical execution. Together, these results show robust instruction following, strong generalization, and real-world feasibility. We release both the model and dataset to support open, reproducible video-based robot learning. Our website is available at https://www.boyuan.space/large-video-planner/.
Authors: Ping Zhong, Shiyong Meng, Bolei Chen, Tao Zou, Chaoxu Mu, Jianxin Wang
Abstract: Visual Floorplan Localization (FLoc) struggles with severe structural aliasing caused by repetitive minimalist layouts. This occurs because physically distant poses share highly similar visual-geometric features, which degrades spatial separability and angular discriminability. While existing methods attempt to mitigate these ambiguities by relying on costly semantic annotations, the resulting performance gains remain inherently limited. To address the above issues, we propose DisCo-FLoc, a semantic-free method for visual-geometric Contrastive Disambiguation. First, we introduce a depth-aware Ray Regression Predictor (RRP) that serves as a dense-to-ray geometric projector. By explicitly suppressing visual clutter along the vertical dimension, RRP projects monocular RGB images into 2D ray primitives, which are matched with floorplans to produce geometry-aware FLoc candidates. Second, to resolve the remaining ambiguity among these candidates, we propose a spatially perturbed contrastive objective to align RGB images with local floorplan structures and formulate a visual-geometric compatibility function. In particular, we meticulously construct positive and negative samples at both positional and directional levels through $SE(2)$ pose perturbations for contrastive learning, effectively achieving pose smoothness, spatial separability, and angular discriminability. The compatibility function enables DisCo-FLoc to disambiguate FLoc by using richer visual context beyond pure geometric layouts, without requiring any semantic annotations. Extensive experiments on two challenging visual FLoc benchmarks demonstrate that DisCo-FLoc significantly outperforms state-of-the-art semantic-based methods, especially narrowing the performance gap between positional and directional FLoc accuracy.
Authors: Bostan Khan, Masoud Daneshtalab
Abstract: Deploying federated learning across heterogeneous IoT device fleets requires tailored neural network architectures for each device class, yet existing Federated Neural Architecture Search (FedNAS) methods suffer from unguided supernet training and prohibitively costly post-training search pipelines that demand over 20 GPU-hours per deployment target. We introduce DeepFedNAS, a two-phase framework built on a multi-objective fitness function that synthesizes information-theoretic network metrics with architectural heuristics. In the first phase, Federated Pareto Optimal Supernet Training replaces random subnet sampling with a pre-computed cache of elite, high-fitness architectures, yielding a superior supernet. In the second phase, a Predictor-Free Search uses this fitness function as a zero-cost accuracy proxy, discovering hardware-optimized subnets in ~20 seconds, a ~61x speedup over the baseline pipeline. Experiments on CIFAR-10, CIFAR-100, and CINIC-10 demonstrate state-of-the-art accuracy (up to +1.21% on CIFAR-100), a 2.8x reduction in per-round transmission size, and robust performance under extreme non-IID conditions ({\alpha} = 0.1), making DeepFedNAS practical for scalable, communication-constrained IoT federations. Source code: https://github.com/bostankhan6/DeepFedNAS
Authors: Anderson de Andrade, Alon Harell, Ivan V. Baji\'c
Abstract: Many computer vision tasks share substantial overlapping information, yet conventional codecs tend to ignore this, leading to redundant and inefficient representations. The Gray-Wyner network, a classical concept from information theory, offers a principled framework for separating common and task-specific information. Inspired by this idea, we develop a learnable three-channel codec that disentangles shared information from task-specific details across multiple vision tasks. We characterize the limits of this approach through the notion of lossy common information, and propose an optimization objective that balances inherent tradeoffs in learning such representations. Through comparisons of three codec architectures on two-task scenarios spanning six vision benchmarks, we demonstrate that our approach substantially reduces redundancy and consistently outperforms independent coding. These results highlight the practical value of revisiting Gray-Wyner theory in modern machine learning contexts, bridging classic information theory with task-driven representation learning.
Authors: Meng Lou, Yunxiang Fu, Yizhou Yu
Abstract: Continual learning, especially class-incremental learning (CIL), on the basis of a pre-trained model (PTM) has garnered substantial research interest in recent years. However, how to effectively learn both discriminative and comprehensive feature representations while maintaining stability and plasticity over very long task sequences remains an open problem. We propose CaRE, a scalable {C}ontinual Le{a}rner with efficient Bi-Level {R}outing Mixture-of-{E}xperts (BR-MoE). The core idea of BR-MoE is a bi-level routing mechanism: a router selection stage that dynamically activates relevant task-specific routers, followed by an expert routing phase that dynamically activates and aggregates experts, aiming to inject discriminative and comprehensive representations into every intermediate network layer. On the other hand, we introduce a challenging dataset, OmniBenchmark-1K, for CIL performance evaluation on very long task sequences with hundreds of tasks. Extensive experiments show that CaRE demonstrates leading performance across a variety of datasets and task settings, including commonly used CIL datasets with classical CIL settings (e.g., 5-20 tasks). To the best of our knowledge, CaRE is the first continual learner that scales to very long task sequences (ranging from 100 to over 300 non-overlapping tasks), while outperforming all baselines by a large margin on such task sequences. We hope that this work will inspire further research into continual learning over extremely long task sequences. Code and dataset are publicly released at https://github.com/LMMMEng/CaRE.
Authors: Phil Blandfort, Tushar Karayil, Alex McKenzie, Urja Pawar, Robert Graham, Dmitrii Krasheninnikov
Abstract: Moral benchmarks for LLMs typically score models on context-free prompts, implicitly treating the measured choice rate as stable. We test this assumption with a direction-flipped influence audit: for each scenario, we compare a baseline prompt with matched cues steering toward option A or option B. Across a trolley-problem-style moral triage task, BBQ, and DailyDilemmas, and across five LLM families with and without reasoning, short contextual cues shift per-condition choice rates by 12-18 percentage points on average. These shifts reveal structure that baseline scores miss: roughly 40% of baseline-neutral triage and BBQ conditions exhibit directional asymmetry under influence, and a meaningful share of significant effects backfire, moving opposite the cue's intended direction. In follow-up probes, models often recognize the cue while denying that it affected their choice. Among significant backfire trials, this stated-vs.-revealed inconsistency appears in 78% of cases. Reasoning does not eliminate contextual sensitivity but reshapes it: social-pressure cues such as user preference and emotional appeal weaken across benchmarks, while few-shot demonstrations strengthen sharply on both triage and BBQ. We recommend direction-flipped influence pairs as a standard complement to context-free moral-bias evaluation, and release the harness and data to make such audits routine.
Authors: Dylan Albrecht, Sarah Thompson, N. Tobias Jacobson, Ryan Jock
Abstract: The exchange interaction is a foundational building block for the operation of spin-based quantum processors. Extracting the exchange interaction coefficient $J(\mathbf{V})$, as a function of gate electrode voltages, is important for understanding disorder, faithfully simulating device performance, and operating spin qubits with high fidelity. Typical coherent measurements of exchange in spin qubit devices yield a modulated cosine of an accumulated phase, which in turn is the time integral of exchange. As such, extracting $J(\mathbf{V})$ from experimental data is difficult due to the ambiguity of inverting a cosine, the sensitivity to noise when unwrapping phase, as well as the problem of inverting the integral. As a step toward obtaining $J(\mathbf{V})$, we tackle the first two challenges to reveal the accumulated phase, $\phi(\mathbf{V})$. We incorporate techniques from a wide range of fields to robustly extract and model a 3D phase volume for spin qubit devices from a sequence of 2D measurements. In particular, we present a measurement technique to obtain the wrapped phase, as done in phase-shifting digital holography, and utilize the max-flow/min-cut phase unwrapping method (PUMA) to unwrap the phase in 3D voltage space. We show this method is robust to the minimal observed drift in the device, which we confirm by increasing scan resolution. Upon building a model of the extracted phase, we optimize over the model to locate a minimal-gradient $\pi$ exchange pulse point in voltage space. Our measurement protocol may provide detailed information useful for understanding the origins of device variability governing device yield, enable calibrating device models to specific devices during operation for more sophisticated error attribution, and enable a systematic optimization of qubit control. We anticipate that the methods presented here may be applicable to other qubit platforms.
Authors: Leonard T. Franz, Sebastian Hoffmann, Tim Weiland, Bernhard Sch\"olkopf, Georg Martius
Abstract: Drifting models have recently gained attention for generating high-quality samples in a single forward pass. During training, they learn a push-forward map by following a vector-valued field, the drift field. We ask whether this procedure is equivalent to optimizing a scalar loss and find that, in general, it is not: drift fields are not conservative and cannot be written as the gradient of any scalar potential. We identify the position-dependent normalization as the source of non-conservatism, with the Gaussian kernel as the unique radial exception. Guided by this, we introduce the sharp kernel $k^\#$ and a sharp-normalized drift field that is conservative for general radial kernels. The resulting vector field is the gradient of a scalar potential that can be optimized directly using stochastic gradient descent. Moreover, the field has the form of a score difference of kernel density estimates, and gives exact equilibrium identifiability. Thus, sharp normalization closes the gap to related literature, such as Wasserstein gradient-flows and denoising score matching, also for non-Gaussian kernels. Empirically, sharp normalization preserves the performance of the original drifting objective, suggesting that the non-conservative flexibility is not required for high-quality generation.
Authors: Tianwei Ye, Yifan Mao, Minwen Liao, Jian Liu, Chunchao Guo, Dazhao Du, Quanxin Shou, Fangqi Zhu, Song Guo
Abstract: Embodied AI and robotic systems increasingly depend on scalable, diverse, and physically grounded 3D content for simulation-based training and real-world deployment. While 3D generative modeling has advanced rapidly, embodied applications impose requirements far beyond visual realism: generated objects must carry kinematic structure and material properties, scenes must support interaction and task execution, and the resulting content must bridge the gap between simulation and reality. This survey reviews 3D generation for embodied AI and organizes the literature around three roles that 3D generation plays in embodied systems. In Data Generator, 3D generation produces simulation-ready objects and assets, including articulated, physically grounded, and deformable content for downstream interaction; in Simulation Environments, it constructs interactive and task-oriented worlds, spanning structure-aware, controllable, and agentic scene generation; and in Sim2Real Bridge, it supports digital twin reconstruction, data augmentation, and synthetic demonstrations for downstream robot learning and real-world transfer. We also show that the field is shifting from visual realism toward interaction readiness, and we identify the main bottlenecks, including limited physical annotations, the gap between geometric quality and physical validity, fragmented evaluation, and the persistent sim-to-real divide, that must be addressed for 3D generation to become a dependable foundation for embodied intelligence. Our project page is at https://3dgen4robot.github.io.
Authors: Haojian Huang, Jiahao Shi, Yinchuan Li, Yingcong Chen
Abstract: Affordance grounding requires identifying where and how an agent should interact in open-world scenes, where actionable regions are often small, occluded, reflective, and visually ambiguous. Recent systems therefore combine multiple skills (e.g., detection, segmentation, interaction-imagination), yet most orchestrate them with fixed pipelines that are poorly matched to per-instance difficulty, offer limited targeted recovery from intermediate errors, and fail to reuse experience from recurring objects. These failures expose a systems problem: test-time grounding must acquire the right evidence, decide whether that evidence is reliable enough to commit, and do so under bounded inference cost without access to labels. We propose Affordance Agent Harness, a closed-loop runtime that unifies heterogeneous skills with an evidence store and cost control, retrieves episodic memories to provide priors for recurring categories, and employs a Router to adaptively select and parameterize skills. An affordance-specific Verifier then gates commitments using self-consistency, cross-scale stability, and evidence sufficiency, triggering targeted retries before a final judge fuses accumulated evidence and trajectories into the prediction. Experiments on multiple affordance benchmarks and difficulty-controlled subsets show a stronger accuracy-cost Pareto frontier than fixed-pipeline baselines, improving grounding quality while reducing average skill calls and latency. Project page: https://tenplusgood.github.io/a-harness-page/.
Authors: Mae Younes, Adnane Boukhayma
Abstract: We present 3D Surface Splatting (3DSS), the first differentiable surface splatting renderer for physically-based inverse rendering from multi-view images. Our central insight is that the surface separation problem at the heart of surface splatting admits a direct formulation in terms of the reconstruction kernels themselves. From this foundation we derive a coverage-based compositing model whose per-layer opacity arises directly from the accumulated Elliptical Weighted Average reconstruction weight, yielding anti-aliased silhouettes and informative visibility gradients at sparsely covered edges. Combined with forward microfacet shading under co-optimized HDR environment lighting and density-aware adaptive refinement, 3DSS jointly recovers shape, spatially-varying BRDF materials, and illumination. Because the optimized representation is a set of oriented surface samples, it bridges natively to mesh-based workflows via surface reconstruction from oriented point cloud methods. We evaluate 3DSS against mesh-based, implicit, and Gaussian-splatting baselines across geometry reconstruction, novel-view synthesis, and novel-illumination relighting.
Authors: Jakub St\k{e}pie\'n, Marcin Mazur, Jacek Tabor, Przemys{\l}aw Spurek
Abstract: Sparse Autoencoders (SAEs) have become an important tool in mechanistic interpretability, helping to analyze internal representations in both Large Language Models (LLMs) and Vision Transformers (ViTs). By decomposing polysemantic activations into sparse sets of monosemantic features, SAEs aim to translate neural network computations into human-understandable concepts. However, common architectures such as TopK SAEs rely on a fixed sparsity level. They enforce the same number of active features (K) across all inputs, ignoring the varying complexity of real-world data. Natural data often lies on manifolds with varying local intrinsic dimensionality, meaning the number of relevant factors can change significantly across samples. This suggests that a fixed sparsity level is not optimal. Simple inputs may require only a few features, while more complex ones need more expressive representations. Using a constant K can therefore introduce noise in simple cases or miss important structure in more complex ones. To address this issue, we propose SoftSAE, a sparse autoencoder with a Dynamic Top-K selection mechanism. Our method uses a differentiable Soft Top-K operator to learn an input-dependent sparsity level k. This allows the model to adjust the number of active features based on the complexity of each input. As a result, the representation better matches the structure of the data, and the explanation length reflects the amount of information in the input. Experimental results confirm that SoftSAE not only finds meaningful features, but also selects the right number of features for each concept. The source code is available at: https://github.com/St0pien/SoftSAE.