Authors: Abhijeet Kumar, Chetan Agarwal, Pronoy B. Neogi, Mayank Goswami
Abstract: This research explores the fusion of graphology and artificial intelligence to quantify psychological stress levels in students by analyzing their handwritten examination scripts. By leveraging Optical Character Recognition and transformer based sentiment analysis models, we present a data driven approach that transcends traditional grading systems, offering deeper insights into cognitive and emotional states during examinations. The system integrates high resolution image processing, TrOCR, and sentiment entropy fusion using RoBERTa based models to generate a numerical Stress Index. Our method achieves robustness through a five model voting mechanism and unsupervised anomaly detection, making it an innovative framework in academic forensics.
Authors: Aswath Muthuselvam, Jeevak Raj S, Mohanaprasad K
Abstract: Road conditions play an important role in our everyday commute. With the proliferating number of vehicles on the road each year, it has become necessary to access the road conditions very frequently, this would ensure that the traffic also flows smoothly. Even the smallest crack in the road could be easily be chipped into a large pothole due to changing surface temperatures of the road and from the force of vehicles riding over it. In this paper, we have addressed how we could better identify these potholes in realtime with the help of onboard sensors in vehicles so that the data could be useful for analysis and better management of potholes on a large scale. For the implementation, we used an SVM classifier to detect potholes, we achieved 98.1% accuracy based on data collected from a local road for about 2 km which had 26 potholes distributed along the road. Code is available at: https://github.com/aswathselvam/Potholes
Authors: Kesong Zheng, Zhi Song, Peizhou Li, Shuyi Yao, Zhenxing Bian
Abstract: Addressing the current lack of a standardized habitat classification system for cultivated land ecosystems, incomplete coverage of habitat types, and the inability of existing models to effectively integrate semantic and texture features-resulting in insufficient segmentation accuracy and blurred boundaries for multi-scale habitats (e.g., large-scale field plots and micro-habitats)-this study developed a comprehensively annotated ultra-high-resolution remote sensing image dataset encompassing 15 categories of cultivated land system habitats. Furthermore, we propose a Dynamic-Weighted Feature Fusion Network (DWFF-Net). The encoder of this model utilizes a frozen-parameter DINOv3 to extract foundational features. By analyzing the relationships between different category images and feature maps, we introduce a data-level adaptive dynamic weighting strategy for feature fusion. The decoder incorporates a dynamic weight computation network to achieve thorough integration of multi-layer features, and a hybrid loss function is adopted to optimize model training. Experimental results on the constructed dataset demonstrate that the proposed model achieves a mean Intersection over Union (mIoU) of 0.6979 and an F1-score of 0.8049, outperforming the baseline network by 0.021 and 0.0161, respectively. Ablation studies further confirm the complementary nature of multi-layer feature fusion, which effectively improves the IoU for micro-habitat categories such as field ridges. This study establishes a habitat identification framework for cultivated land systems based on adaptive multi-layer feature fusion, enabling sub-meter precision habitat mapping at a low cost and providing robust technical support for fine-grained habitat monitoring in cultivated landscapes.
Authors: Ziyuan Gao
Abstract: Medical image segmentation requires large annotated datasets, creating a significant bottleneck for clinical applications. While few-shot segmentation methods can learn from minimal examples, existing approaches demonstrate suboptimal performance in precise boundary delineation for medical images, particularly when anatomically similar regions appear without sufficient spatial context. We propose AGENet (Adaptive Geodesic Edge-aware Network), a novel framework that incorporates spatial relationships through edge-aware geodesic distance learning. Our key insight is that medical structures follow predictable geometric patterns that can guide prototype extraction even with limited training data. Unlike methods relying on complex architectural components or heavy neural networks, our approach leverages computationally lightweight geometric modeling. The framework combines three main components: (1) An edge-aware geodesic distance learning module that respects anatomical boundaries through iterative Fast Marching refinement, (2) adaptive prototype extraction that captures both global structure and local boundary details via spatially-weighted aggregation, and (3) adaptive parameter learning that automatically adjusts to different organ characteristics. Extensive experiments across diverse medical imaging datasets demonstrate improvements over state-of-the-art methods. Notably, our method reduces boundary errors compared to existing approaches while maintaining computational efficiency, making it highly suitable for clinical applications requiring precise segmentation with limited annotated data.
Authors: Jiahui Wang, Haiyue Zhu, Haoren Guo, Abdullah Al Mamun, Cheng Xiang, Tong Heng Lee
Abstract: Recent approaches for few-shot 3D point cloud semantic segmentation typically require a two-stage learning process, i.e., a pre-training stage followed by a few-shot training stage. While effective, these methods face overreliance on pre-training, which hinders model flexibility and adaptability. Some models tried to avoid pre-training yet failed to capture ample information. In addition, current approaches focus on visual information in the support set and neglect or do not fully exploit other useful data, such as textual annotations. This inadequate utilization of support information impairs the performance of the model and restricts its zero-shot ability. To address these limitations, we present a novel pre-training-free network, named Efficient Point Cloud Semantic Segmentation for Few- and Zero-shot scenarios. Our EPSegFZ incorporates three key components. A Prototype-Enhanced Registers Attention (ProERA) module and a Dual Relative Positional Encoding (DRPE)-based cross-attention mechanism for improved feature extraction and accurate query-prototype correspondence construction without pre-training. A Language-Guided Prototype Embedding (LGPE) module that effectively leverages textual information from the support set to improve few-shot performance and enable zero-shot inference. Extensive experiments show that our method outperforms the state-of-the-art method by 5.68% and 3.82% on the S3DIS and ScanNet benchmarks, respectively.
Authors: Lian He, Meng Liu, Qilang Ye, Yu Zhou, Xiang Deng, Gangyi Ding
Abstract: Understanding 3D scene-level affordances from natural language instructions is essential for enabling embodied agents to interact meaningfully in complex environments. However, this task remains challenging due to the need for semantic reasoning and spatial grounding. Existing methods mainly focus on object-level affordances or merely lift 2D predictions to 3D, neglecting rich geometric structure information in point clouds and incurring high computational costs. To address these limitations, we introduce Task-Aware 3D Scene-level Affordance segmentation (TASA), a novel geometry-optimized framework that jointly leverages 2D semantic cues and 3D geometric reasoning in a coarse-to-fine manner. To improve the affordance detection efficiency, TASA features a task-aware 2D affordance detection module to identify manipulable points from language and visual inputs, guiding the selection of task-relevant views. To fully exploit 3D geometric information, a 3D affordance refinement module is proposed to integrate 2D semantic priors with local 3D geometry, resulting in accurate and spatially coherent 3D affordance masks. Experiments on SceneFun3D demonstrate that TASA significantly outperforms the baselines in both accuracy and efficiency in scene-level affordance segmentation.
Authors: Pouya Shiri, Amirali Baniasadi
Abstract: Capsule Network (CapsNet) classifier has several advantages over CNNs, including better detection of images containing overlapping categories and higher accuracy on transformed images. Despite the advantages, CapsNet is slow due to its different structure. In addition, CapsNet is resource-hungry, includes many parameters and lags in accuracy compared to CNNs. In this work, we propose LE-CapsNet as a light, enhanced and more accurate variant of CapsNet. Using 3.8M weights, LECapsNet obtains 76.73% accuracy on the CIFAR-10 dataset while performing inference 4x faster than CapsNet. In addition, our proposed network is more robust at detecting images with affine transformations compared to CapsNet. We achieve 94.3% accuracy on the AffNIST dataset (compared to CapsNet 90.52%).
Authors: Zhou Xu, Qi Wang, Yuxiao Yang, Luyuan Zhang, Zhang Liang, Yang Li
Abstract: Score Distillation Sampling (SDS) enables 3D asset generation by distilling priors from pretrained 2D text-to-image diffusion models, but vanilla SDS suffers from over-saturation and over-smoothing. To mitigate this issue, recent variants have incorporated negative prompts. However, these methods face a critical trade-off: limited texture optimization, or significant texture gains with shape distortion. In this work, we first conduct a systematic analysis and reveal that this trade-off is fundamentally governed by the utilization of the negative prompts, where Target Negative Prompts (TNP) that embed target information in the negative prompts dramatically enhancing texture realism and fidelity but inducing shape distortions. Informed by this key insight, we introduce the Target-Balanced Score Distillation (TBSD). It formulates generation as a multi-objective optimization problem and introduces an adaptive strategy that effectively resolves the aforementioned trade-off. Extensive experiments demonstrate that TBSD significantly outperforms existing state-of-the-art methods, yielding 3D assets with high-fidelity textures and geometrically accurate shape.
Authors: Sudhakar Sah, Nikhil Chabbra, Matthieu Durnerin
Abstract: Deep Convolutional Neural Networks (CNNs) are increasingly difficult to deploy on microcontrollers (MCUs) and lightweight NPUs (Neural Processing Units) due to their growing size and compute demands. Low-rank tensor decomposition, such as Tucker factorization, is a promising way to reduce parameters and operations with reasonable accuracy loss. However, existing approaches select ranks locally and often ignore global trade-offs between compression and accuracy. We introduce CompressNAS, a MicroNAS-inspired framework that treats rank selection as a global search problem. CompressNAS employs a fast accuracy estimator to evaluate candidate decompositions, enabling efficient yet exhaustive rank exploration under memory and accuracy constraints. In ImageNet, CompressNAS compresses ResNet-18 by 8x with less than 4% accuracy drop; on COCO, we achieve 2x compression of YOLOv5s without any accuracy drop and 2x compression of YOLOv5n with a 2.5% drop. Finally, we present a new family of compressed models, STResNet, with competitive performance compared to other efficient models.
Authors: Jiao Chen, Haoyi Wang, Jianhua Tang, Junyi Wang
Abstract: Low-altitude Unmanned Aerial Vehicle (UAV) networks rely on robust semantic segmentation as a foundational enabler for distributed sensing-communication-control co-design across heterogeneous agents within the network. However, segmentation foundation models deteriorate quickly under weather, lighting, and viewpoint drift. Resource-limited UAVs cannot run gradient-based test-time adaptation, while resource-massive UAVs adapt independently, wasting shared experience. To address these challenges, we propose AdaptFly, a prompt-guided test-time adaptation framework that adjusts segmentation models without weight updates. AdaptFly features two complementary adaptation modes. For resource-limited UAVs, it employs lightweight token-prompt retrieval from a shared global memory. For resource-massive UAVs, it uses gradient-free sparse visual prompt optimization via Covariance Matrix Adaptation Evolution Strategy. An activation-statistic detector triggers adaptation, while cross-UAV knowledge pool consolidates prompt knowledge and enables fleet-wide collaboration with negligible bandwidth overhead. Extensive experiments on UAVid and VDD benchmarks, along with real-world UAV deployments under diverse weather conditions, demonstrate that AdaptFly significantly improves segmentation accuracy and robustness over static models and state-of-the-art TTA baselines. The results highlight a practical path to resilient, communication-efficient perception in the emerging low-altitude economy.
Authors: Zekai Shi, Zhixi Cai, Kalin Stefanov
Abstract: Typically, children start to learn their first words between 6 and 9 months, linking spoken utterances to their visual referents. Without prior knowledge, a word encountered for the first time can be interpreted in countless ways; it might refer to any of the objects in the environment, their components, or attributes. Using longitudinal, egocentric, and ecologically valid data from the experience of one child, in this work, we propose a self-supervised and biologically plausible strategy to learn strong visual representations. Our masked autoencoder-based visual backbone incorporates knowledge about the blind spot in human eyes to define a novel masking strategy. This mask and reconstruct approach attempts to mimic the way the human brain fills the gaps in the eyes' field of view. This represents a significant shift from standard random masking strategies, which are difficult to justify from a biological perspective. The pretrained encoder is utilized in a contrastive learning-based video-text model capable of acquiring word-referent mappings. Extensive evaluation suggests that the proposed biologically plausible masking strategy is at least as effective as random masking for learning word-referent mappings from cross-situational and temporally extended episodes.
Authors: Yongjun Xiao, Dian Meng, Xinlei Huang, Yanran Liu, Shiwei Ruan, Ziyue Qiao, Xubin Zheng
Abstract: Effectively modeling multimodal spatial omics data is critical for understanding tissue complexity and underlying biological mechanisms. While spatial transcriptomics, proteomics, and epigenomics capture molecular features, they lack pathological morphological context. Integrating these omics with histopathological images is therefore essential for comprehensive disease tissue analysis. However, substantial heterogeneity across omics, imaging, and spatial modalities poses significant challenges. Naive fusion of semantically distinct sources often leads to ambiguous representations. Additionally, the resolution mismatch between high-resolution histology images and lower-resolution sequencing spots complicates spatial alignment. Biological perturbations during sample preparation further distort modality-specific signals, hindering accurate integration. To address these challenges, we propose Graph-guided Representation of Omics and Vision with Expert Regulation for Adaptive Spatial Multi-omics Fusion (GROVER), a novel framework for adaptive integration of spatial multi-omics data. GROVER leverages a Graph Convolutional Network encoder based on Kolmogorov-Arnold Networks to capture the nonlinear dependencies between each modality and its associated spatial structure, thereby producing expressive, modality-specific embeddings. To align these representations, we introduce a spot-feature-pair contrastive learning strategy that explicitly optimizes the correspondence across modalities at each spot. Furthermore, we design a dynamic expert routing mechanism that adaptively selects informative modalities for each spot while suppressing noisy or low-quality inputs. Experiments on real-world spatial omics datasets demonstrate that GROVER outperforms state-of-the-art baselines, providing a robust and reliable solution for multimodal integration.
Authors: Aditya Mehta, Swarnim Chaudhary, Pratik Narang, Jagat Sesh Challa
Abstract: Modern generative and diffusion models produce highly realistic images that can mislead human perception and even sophisticated automated detection systems. Most detection methods operate in RGB space and thus analyze only three spectral channels. We propose HSI-Detect, a two-stage pipeline that reconstructs a 31-channel hyperspectral image from a standard RGB input and performs detection in the hyperspectral domain. Expanding the input representation into denser spectral bands amplifies manipulation artifacts that are often weak or invisible in the RGB domain, particularly in specific frequency bands. We evaluate HSI-Detect across FaceForensics++ dataset and show the consistent improvements over RGB-only baselines, illustrating the promise of spectral-domain mapping for Deepfake detection.
Authors: Yonatan Sverdlov, Eitan Rosen, Nadav Dym
Abstract: Many neural networks for point clouds are, by design, invariant to the symmetries of this datatype: permutations and rigid motions. The purpose of this paper is to examine whether such networks preserve natural symmetry aware distances on the point cloud spaces, through the notion of bi-Lipschitz equivalence. This inquiry is motivated by recent work in the Equivariant learning literature which highlights the advantages of bi-Lipschitz models in other scenarios. We consider two symmetry aware metrics on point clouds: (a) The Procrustes Matching (PM) metric and (b) Hard Gromov Wasserstien distances. We show that these two distances themselves are not bi-Lipschitz equivalent, and as a corollary deduce that popular invariant networks for point clouds are not bi-Lipschitz with respect to the PM metric. We then show how these networks can be modified so that they do obtain bi-Lipschitz guarantees. Finally, we provide initial experiments showing the advantage of the proposed bi-Lipschitz model over standard invariant models, for the tasks of finding correspondences between 3D point clouds.
Authors: Sanchit Sinha, Guangzhi Xiong, Zhenghao He, Aidong Zhang
Abstract: Modern vision-language models (VLMs) deliver impressive predictive accuracy yet offer little insight into 'why' a decision is reached, frequently hallucinating facts, particularly when encountering out-of-distribution data. Neurosymbolic frameworks address this by pairing black-box perception with interpretable symbolic reasoning, but current methods extract their symbols solely from task labels, leaving them weakly grounded in the underlying visual data. In this paper, we introduce a multi-agent system - Concept-RuleNet that reinstates visual grounding while retaining transparent reasoning. Specifically, a multimodal concept generator first mines discriminative visual concepts directly from a representative subset of training images. Next, these visual concepts are utilized to condition symbol discovery, anchoring the generations in real image statistics and mitigating label bias. Subsequently, symbols are composed into executable first-order rules by a large language model reasoner agent - yielding interpretable neurosymbolic rules. Finally, during inference, a vision verifier agent quantifies the degree of presence of each symbol and triggers rule execution in tandem with outputs of black-box neural models, predictions with explicit reasoning pathways. Experiments on five benchmarks, including two challenging medical-imaging tasks and three underrepresented natural-image datasets, show that our system augments state-of-the-art neurosymbolic baselines by an average of 5% while also reducing the occurrence of hallucinated symbols in rules by up to 50%.
Authors: Stanislav Selitskiy
Abstract: A novel Transformer variation architecture is proposed in the implicit sparse style. Unlike "traditional" Transformers, instead of attention to sequential or batch entities in their entirety of whole dimensionality, in the proposed Batch Transformers, attention to the "important" dimensions (primary components) is implemented. In such a way, the "important" dimensions or feature selection allows for a significant reduction of the bottleneck size in the encoder-decoder ANN architectures. The proposed architecture is tested on the synthetic image generation for the face recognition task in the case of the makeup and occlusion data set, allowing for increased variability of the limited original data set.
Authors: Hossein Mohebbi, Mohammed Abdulrahman, Yanting Miao, Pascal Poupart, Suraj Kothawade
Abstract: Recent advances in text-to-image generation have produced strong single-shot models, yet no individual system reliably executes the long, compositional prompts typical of creative workflows. We introduce Image-POSER, a reflective reinforcement learning framework that (i) orchestrates a diverse registry of pretrained text-to-image and image-to-image experts, (ii) handles long-form prompts end-to-end through dynamic task decomposition, and (iii) supervises alignment at each step via structured feedback from a vision-language model critic. By casting image synthesis and editing as a Markov Decision Process, we learn non-trivial expert pipelines that adaptively combine strengths across models. Experiments show that Image-POSER outperforms baselines, including frontier models, across industry-standard and custom benchmarks in alignment, fidelity, and aesthetics, and is consistently preferred in human evaluations. These results highlight that reinforcement learning can endow AI systems with the capacity to autonomously decompose, reorder, and combine visual models, moving towards general-purpose visual assistants.
Authors: Zhongping Dong, Pengyang Yu, Shuangjian Li, Liming Chen, Mohand Tahar Kechadi
Abstract: Accurate single-object tracking and short-term motion forecasting remain challenging under occlusion, scale variation, and temporal drift, which disrupt the temporal coherence required for real-time perception. We introduce \textbf{SOTFormer}, a minimal constant-memory temporal transformer that unifies object detection, tracking, and short-horizon trajectory prediction within a single end-to-end framework. Unlike prior models with recurrent or stacked temporal encoders, SOTFormer achieves stable identity propagation through a ground-truth-primed memory and a burn-in anchor loss that explicitly stabilizes initialization. A single lightweight temporal-attention layer refines embeddings across frames, enabling real-time inference with fixed GPU memory. On the Mini-LaSOT (20%) benchmark, SOTFormer attains 76.3 AUC and 53.7 FPS (AMP, 4.3 GB VRAM), outperforming transformer baselines such as TrackFormer and MOTRv2 under fast motion, scale change, and occlusion.
Authors: Fatemeh Elhambakhsh, Gaurav Ameta, Aditi Roy, Hyunwoong Ko
Abstract: Machining process planning (MP) is inherently complex due to structural and geometrical dependencies among part features and machining operations. A key challenge lies in capturing dynamic interdependencies that evolve with distinct part geometries as operations are performed. Machine learning has been applied to address challenges in MP, such as operation selection and machining sequence prediction. Dynamic graph learning (DGL) has been widely used to model dynamic systems, thanks to its ability to integrate spatio-temporal relationships. However, in MP, while existing DGL approaches can capture these dependencies, they fail to incorporate three-dimensional (3D) geometric information of parts and thus lack domain awareness in predicting machining operation sequences. To address this limitation, we propose MP-GFormer, a 3D-geometry-aware dynamic graph transformer that integrates evolving 3D geometric representations into DGL through an attention mechanism to predict machining operation sequences. Our approach leverages StereoLithography surface meshes representing the 3D geometry of a part after each machining operation, with the boundary representation method used for the initial 3D designs. We evaluate MP-GFormer on a synthesized dataset and demonstrate that the method achieves improvements of 24\% and 36\% in accuracy for main and sub-operation predictions, respectively, compared to state-of-the-art approaches.
Authors: Wei-Jia Chen, Min-Yen Tsai, Cheng-Yi Lee, Chia-Mu Yu
Abstract: The rapid proliferation of pretrained models and open repositories has made model merging a convenient yet risky practice, allowing free-riders to combine fine-tuned models into a new multi-capability model without authorization. Such unauthorized model merging not only violates intellectual property rights but also undermines model ownership and accountability. To address this issue, we present MergeGuard, a proactive dual-stage weight protection framework that disrupts merging compatibility while maintaining task fidelity. In the first stage, we redistribute task-relevant information across layers via L2-regularized optimization, ensuring that important gradients are evenly dispersed. In the second stage, we inject structured perturbations to misalign task subspaces, breaking curvature compatibility in the loss landscape. Together, these stages reshape the model's parameter geometry such that merged models collapse into destructive interference while the protected model remains fully functional. Extensive experiments on both vision (ViT-L-14) and language (Llama2, Gemma2, Mistral) models demonstrate that MergeGuard reduces merged model accuracy by up to 90% with less than 1.5% performance loss on the protected model.
Authors: Muzammal Shafique, Nasir Rahim, Jamil Ahmad, Mohammad Siadat, Khalid Malik, Ghaus Malik
Abstract: Segmentation of medical images constitutes an essential component of medical image analysis, providing the foundation for precise diagnosis and efficient therapeutic interventions in clinical practices. Despite substantial progress, most segmentation models do not explicitly encode boundary information; as a result, making boundary preservation a persistent challenge in medical image segmentation. To address this challenge, we introduce FocusSDF, a novel loss function based on the signed distance functions (SDFs), which redirects the network to concentrate on boundary regions by adaptively assigning higher weights to pixels closer to the lesion or organ boundary, effectively making it boundary aware. To rigorously validate FocusSDF, we perform extensive evaluations against five state-of-the-art medical image segmentation models, including the foundation model MedSAM, using four distance-based loss functions across diverse datasets covering cerebral aneurysm, stroke, liver, and breast tumor segmentation tasks spanning multiple imaging modalities. The experimental results consistently demonstrate the superior performance of FocusSDF over existing distance transform based loss functions.
Authors: Simon Durand, Samuel Foucher, Alexandre Delplanque, Jo\"elle Taillon, J\'er\^ome Th\'eau
Abstract: Accurate population estimates are essential for wildlife management, providing critical insights into species abundance and distribution. Traditional survey methods, including visual aerial counts and GNSS telemetry tracking, are widely used to monitor muskox populations in Arctic regions. These approaches are resource intensive and constrained by logistical challenges. Advances in remote sensing, artificial intelligence, and high resolution aerial imagery offer promising alternatives for wildlife detection. Yet, the effectiveness of deep learning object detection models (ODMs) is often limited by small datasets, making it challenging to train robust ODMs for sparsely distributed species like muskoxen. This study investigates the integration of synthetic imagery (SI) to supplement limited training data and improve muskox detection in zero shot (ZS) and few-shot (FS) settings. We compared a baseline model trained on real imagery with 5 ZS and 5 FS models that incorporated progressively more SI in the training set. For the ZS models, where no real images were included in the training set, adding SI improved detection performance. As more SI were added, performance in precision, recall and F1 score increased, but eventually plateaued, suggesting diminishing returns when SI exceeded 100% of the baseline model training dataset. For FS models, combining real and SI led to better recall and slightly higher overall accuracy compared to using real images alone, though these improvements were not statistically significant. Our findings demonstrate the potential of SI to train accurate ODMs when data is scarce, offering important perspectives for wildlife monitoring by enabling rare or inaccessible species to be monitored and to increase monitoring frequency. This approach could be used to initiate ODMs without real data and refine it as real images are acquired over time.
Authors: Camila Machado de Araujo, Egon P. B. S. Borges, Ricardo Marcelo Canteiro Grangeiro, Allan Pinto
Abstract: High-resolution volumetric imaging techniques, such as X-ray tomography and advanced microscopy, generate increasingly large datasets that challenge existing tools for efficient processing, segmentation, and interactive exploration. This work introduces new capabilities to Annotat3D through Harpia, a new CUDA-based processing library designed to support scalable, interactive segmentation workflows for large 3D datasets in high-performance computing (HPC) and remote-access environments. Harpia features strict memory control, native chunked execution, and a suite of GPU-accelerated filtering, annotation, and quantification tools, enabling reliable operation on datasets exceeding single-GPU memory capacity. Experimental results demonstrate significant improvements in processing speed, memory efficiency, and scalability compared to widely used frameworks such as NVIDIA cuCIM and scikit-image. The system's interactive, human-in-the-loop interface, combined with efficient GPU resource management, makes it particularly suitable for collaborative scientific imaging workflows in shared HPC infrastructures.
Authors: Arnav Singhvi, Vasiliki Bikia, Asad Aali, Akshay Chaudhari, Roxana Daneshjou
Abstract: Vision-language foundation models (VLMs) show promise for diverse imaging tasks but often underperform on medical benchmarks. Prior efforts to improve performance include model finetuning, which requires large domain-specific datasets and significant compute, or manual prompt engineering, which is hard to generalize and often inaccessible to medical institutions seeking to deploy these tools. These challenges motivate interest in approaches that draw on a model's embedded knowledge while abstracting away dependence on human-designed prompts to enable scalable, weight-agnostic performance improvements. To explore this, we adapt the Declarative Self-improving Python (DSPy) framework for structured automated prompt optimization in medical vision-language systems through a comprehensive, formal evaluation. We implement prompting pipelines for five medical imaging tasks across radiology, gastroenterology, and dermatology, evaluating 10 open-source VLMs with four prompt optimization techniques. Optimized pipelines achieved a median relative improvement of 53% over zero-shot prompting baselines, with the largest gains ranging from 300% to 3,400% on tasks where zero-shot performance is low. These results highlight the substantial potential of applying automated prompt optimization to medical AI systems, demonstrating significant gains for vision-based applications requiring accurate clinical image interpretation. By reducing dependence on prompt design to elicit intended outputs, these techniques allow clinicians to focus on patient care and clinical decision-making. Furthermore, our experiments offer scalability and preserve data privacy, demonstrating performance improvement on open-source VLMs. We publicly release our evaluation pipelines to support reproducible research on specialized medical tasks, available at https://github.com/DaneshjouLab/prompt-triage-lab.
Authors: Afifa Khaled, Ebrahim Hamid Sumiea
Abstract: Medical imaging and multi-modal clinical settings often face the challange of missing modality in their diagnostic pipelines. Existing imputation methods either lack representational capacity or are computationally expensive. We propose PI-NAIM, a novel dual-path architecture that dynamically routes samples to optimized imputation approaches based on missingness complexity. Our framework integrates: (1) intelligent path routing that directs low missingness samples to efficient statistical imputation (MICE) and complex patterns to powerful neural networks (GAIN with temporal analysis); (2) cross-path attention fusion that leverages missingness-aware embeddings to intelligently combine both branches; and (3) end-to-end joint optimization of imputation accuracy and downstream task performance. Extensive experiments on MIMIC-III and multimodal benchmarks demonstrate state-of-the-art performance, achieving RMSE of 0.108 (vs. baselines' 0.119-0.152) and substantial gains in downstream tasks with an AUROC of 0.812 for mortality prediction. PI-NAIM's modular design enables seamless integration into vision pipelines handling incomplete sensor measurements, missing modalities, or corrupted inputs, providing a unified solution for real-world scenario. The code is publicly available at https://github.com/AfifaKhaled/PI-NAIM-Path-Integrated-Neural-Adaptive-Imputation-Model
URLs: https://github.com/AfifaKhaled/PI-NAIM-Path-Integrated-Neural-Adaptive-Imputation-Model
Authors: Siyou Li, Huanan Wu, Juexi Shao, Yinghao Ma, Yujian Gan, Yihao Luo, Yuwei Wang, Dong Nie, Lu Wang, Wengqing Wu, Le Zhang, Massimo Poesio, Juntao Yu
Abstract: Despite the recent advances in the video understanding ability of multimodal large language models (MLLMs), long video understanding remains a challenge. One of the main issues is that the number of vision tokens grows linearly with video length, which causes an explosion in attention cost, memory, and latency. To solve this challenge, we present Query-aware Token Selector (\textbf{QTSplus}), a lightweight yet powerful visual token selection module that serves as an information gate between the vision encoder and LLMs. Given a text query and video tokens, QTSplus dynamically selects the most important visual evidence for the input text query by (i) scoring visual tokens via cross-attention, (ii) \emph{predicting} an instance-specific retention budget based on the complexity of the query, and (iii) \emph{selecting} Top-$n$ tokens with a differentiable straight-through estimator during training and a hard gate at inference. Furthermore, a small re-encoder preserves temporal order using absolute time information, enabling second-level localization while maintaining global coverage. Integrated into Qwen2.5-VL, QTSplus compresses the vision stream by up to \textbf{89\%} and reduces end-to-end latency by \textbf{28\%} on long videos. The evaluation on eight long video understanding benchmarks shows near-parity accuracy overall when compared with the original Qwen models and outperforms the original model by \textbf{+20.5} and \textbf{+5.6} points respectively on TempCompass direction and order accuracies. These results show that QTSplus is an effective, general mechanism for scaling MLLMs to real-world long-video scenarios while preserving task-relevant evidence. We will make all code, data, and trained models' weights publicly available.
Authors: Ling Wang, Yunfan Lu, Wenzong Ma, Huizai Yao, Pengteng Li, Hui Xiong
Abstract: Clear imaging under hazy conditions is a critical task. Prior-based and neural methods have improved results. However, they operate on RGB frames, which suffer from limited dynamic range. Therefore, dehazing remains ill-posed and can erase structure and illumination details. To address this, we use event cameras for dehazing for the \textbf{first time}. Event cameras offer much higher HDR ($120 dBvs.60 dB$) and microsecond latency, therefore they suit hazy scenes. In practice, transferring HDR cues from events to frames is hard because real paired data are scarce. To tackle this, we propose an event-guided diffusion model that utilizes the strong generative priors of diffusion models to reconstruct clear images from hazy inputs by effectively transferring HDR information from events. Specifically, we design an event-guided module that maps sparse HDR event features, \textit{e.g.,} edges, corners, into the diffusion latent space. This clear conditioning provides precise structural guidance during generation, improves visual realism, and reduces semantic drift. For real-world evaluation, we collect a drone dataset in heavy haze (AQI = 341) with synchronized RGB and event sensors. Experiments on two benchmarks and our dataset achieve state-of-the-art results.
Authors: Leonardi Melo, Lu\'is Gustavo, Dimmy Magalh\~aes, Lucciani Vieira, Mauro Ara\'ujo
Abstract: This study presents a comparative analysis of three U-Net-based architectures for semantic segmentation of rock art petroglyphs from Brazilian archaeological sites. The investigated architectures were: (1) BEGL-UNet with Border-Enhanced Gaussian Loss function; (2) Attention-Residual BEGL-UNet, incorporating residual blocks and gated attention mechanisms; and (3) Spatial Channel Attention BEGL-UNet, which employs spatial-channel attention modules based on Convolutional Block Attention Module. All implementations employed the BEGL loss function combining binary cross-entropy with Gaussian edge enhancement. Experiments were conducted on images from the Po\c{c}o da Bebidinha Archaeological Complex, Piau\'i, Brazil, using 5-fold cross-validation. Among the architectures, Attention-Residual BEGL-UNet achieved the best overall performance with Dice Score of 0.710, validation loss of 0.067, and highest recall of 0.854. Spatial Channel Attention BEGL-UNet obtained comparable performance with DSC of 0.707 and recall of 0.857. The baseline BEGL-UNet registered DSC of 0.690. These results demonstrate the effectiveness of attention mechanisms for archaeological heritage digital preservation, with Dice Score improvements of 2.5-2.9% over the baseline.
Authors: Zhenhao Guo, Rachit Saluja, Tianyuan Yao, Quan Liu, Junchao Zhu, Haibo Wang, Daniel Reisenb\"uchler, Yuankai Huo, Benjamin Liechty, David J. Pisapia, Kenji Ikemura, Steven Salvatoree, Surya Seshane, Mert R. Sabuncu, Yihe Yang, Ruining Deng
Abstract: Fine-grained glomerular subtyping is central to kidney biopsy interpretation, but clinically valuable labels are scarce and difficult to obtain. Existing computational pathology approaches instead tend to evaluate coarse diseased classification under full supervision with image-only models, so it remains unclear how vision-language models (VLMs) should be adapted for clinically meaningful subtyping under data constraints. In this work, we model fine-grained glomerular subtyping as a clinically realistic few-shot problem and systematically evaluate both pathology-specialized and general-purpose vision-language models under this setting. We assess not only classification performance (accuracy, AUC, F1) but also the geometry of the learned representations, examining feature alignment between image and text embeddings and the separability of glomerular subtypes. By jointly analyzing shot count, model architecture and domain knowledge, and adaptation strategy, this study provides guidance for future model selection and training under real clinical data constraints. Our results indicate that pathology-specialized vision-language backbones, when paired with the vanilla fine-tuning, are the most effective starting point. Even with only 4-8 labeled examples per glomeruli subtype, these models begin to capture distinctions and show substantial gains in discrimination and calibration, though additional supervision continues to yield incremental improvements. We also find that the discrimination between positive and negative examples is as important as image-text alignment. Overall, our results show that supervision level and adaptation strategy jointly shape both diagnostic performance and multimodal structure, providing guidance for model selection, adaptation strategies, and annotation investment.
Authors: Songsong Zhang, Chuanqi Tang, Hongguang Zhang, Guijian Tang, Minglong Li, Xueqiong Li, Shaowu Yang, Yuanxi Peng, Wenjing Yang, Jing Zhao
Abstract: Identity-Preserving Personalized Generation (IPPG) has advanced film production and artistic creation, yet existing approaches overemphasize facial regions, resulting in outputs dominated by facial close-ups.These methods suffer from weak visual narrativity and poor semantic consistency under complex text prompts, with the core limitation rooted in identity (ID) feature embeddings undermining the semantic expressiveness of generative models. To address these issues, this paper presents an IPPG method that breaks the constraint of facial close-ups, achieving synergistic optimization of identity fidelity and scene semantic creation. Specifically, we design a Dual-Line Inference (DLI) pipeline with identity-semantic separation, resolving the representation conflict between ID and semantics inherent in traditional single-path architectures. Further, we propose an Identity Adaptive Fusion (IdAF) strategy that defers ID-semantic fusion to the noise prediction stage, integrating adaptive attention fusion and noise decision masking to avoid ID embedding interference on semantics without manual masking. Finally, an Identity Aggregation Prepending (IdAP) module is introduced to aggregate ID information and replace random initializations, further enhancing identity preservation. Experimental results validate that our method achieves stable and effective performance in IPPG tasks beyond facial close-ups, enabling efficient generation without manual masking or fine-tuning. As a plug-and-play component, it can be rapidly deployed in existing IPPG frameworks, addressing the over-reliance on facial close-ups, facilitating film-level character-scene creation, and providing richer personalized generation capabilities for related domains.
Authors: Jiaming Liang, Chi-Man Pun
Abstract: Despite their wide application, the vulnerabilities of deep neural networks raise societal concerns. Among them, transformation-based attacks have demonstrated notable success in transfer attacks. However, existing attacks suffer from blind spots in parameter optimization, limiting their full potential. Specifically, (1) prior work generally considers low-iteration settings, yet attacks perform quite differently at higher iterations, so characterizing overall performance based only on low-iteration results is misleading. (2) Existing attacks use uniform parameters for different surrogate models, iterations, and tasks, which greatly impairs transferability. (3) Traditional transformation parameter optimization relies on grid search. For n parameters with m steps each, the complexity is O(mn). Large computational overhead limits further optimization of parameters. To address these limitations, we conduct an empirical study with various transformations as baselines, revealing three dynamic patterns of transferability with respect to parameter strength. We further propose a novel Concentric Decay Model (CDM) to effectively explain these patterns. Building on these insights, we propose an efficient Dynamic Parameter Optimization (DPO) based on the rise-then-fall pattern, reducing the complexity to O(nlogm). Comprehensive experiments on existing transformation-based attacks across different surrogate models, iterations, and tasks demonstrate that our DPO can significantly improve transferability.
Authors: Xinyu He, Botong Zhao, Bingbing Li, Shujing Lyu, Jiwei Shen, Yue Lu
Abstract: Accurate segmentation and measurement of lithography scanning electron microscope (SEM) images are crucial for ensuring precise process control, optimizing device performance, and advancing semiconductor manufacturing yield. Lithography segmentation requires pixel-level delineation of groove contours and consistent performance across diverse pattern geometries and process window. However, existing methods often lack the necessary precision and robustness, limiting their practical applicability. To overcome this challenge, we propose LithoSeg, a coarse-to-fine network tailored for lithography segmentation. In the coarse stage, we introduce a Human-in-the-Loop Bootstrapping scheme for the Segment Anything Model (SAM) to attain robustness with minimal supervision. In the subsequent fine stage, we recast 2D segmentation as 1D regression problem by sampling groove-normal profiles using the coarse mask and performing point-wise refinement with a lightweight MLP. LithoSeg outperforms previous approaches in both segmentation accuracy and metrology precision while requiring less supervision, offering promising prospects for real-world applications.
Authors: Kai-Wen K. Yang, Andrew Bai, Alexandra Bermudez, Yunqi Hong, Zoe Latham, Iris Sloan, Michael Liu, Vishrut Goyal, Cho-Jui Hsieh, Neil Y. C. Lin
Abstract: Deep learning is transforming microscopy, yet models often fail when applied to images from new instruments or acquisition settings. Conventional adversarial domain adaptation (ADDA) retrains entire networks, often disrupting learned semantic representations. Here, we overturn this paradigm by showing that adapting only the earliest convolutional layers, while freezing deeper layers, yields reliable transfer. Building on this principle, we introduce Subnetwork Image Translation ADDA with automatic depth selection (SIT-ADDA-Auto), a self-configuring framework that integrates shallow-layer adversarial alignment with predictive uncertainty to automatically select adaptation depth without target labels. We demonstrate robustness via multi-metric evaluation, blinded expert assessment, and uncertainty-depth ablations. Across exposure and illumination shifts, cross-instrument transfer, and multiple stains, SIT-ADDA improves reconstruction and downstream segmentation over full-encoder adaptation and non-adversarial baselines, with reduced drift of semantic features. Our results provide a design rule for label-free adaptation in microscopy and a recipe for field settings; the code is publicly available.
Authors: Shounak Ray Chaudhuri, Arash Jahangiri, Christopher Paolini
Abstract: Traffic safety analysis at signalized intersections is vital for reducing vehicle and pedestrian collisions, yet traditional crash-based studies are limited by data sparsity and latency. This paper presents a novel multi-camera computer vision framework for real-time safety assessment through Post-Encroachment Time (PET) computation, demonstrated at the intersection of H Street and Broadway in Chula Vista, California. Four synchronized cameras provide continuous visual coverage, with each frame processed on NVIDIA Jetson AGX Xavier devices using YOLOv11 segmentation for vehicle detection. Detected vehicle polygons are transformed into a unified bird's-eye map using homography matrices, enabling alignment across overlapping camera views. A novel pixel-level PET algorithm measures vehicle position without reliance on fixed cells, allowing fine-grained hazard visualization via dynamic heatmaps, accurate to 3.3 sq-cm. Timestamped vehicle and PET data is stored in an SQL database for long-term monitoring. Results over various time intervals demonstrate the framework's ability to identify high-risk regions with sub-second precision and real-time throughput on edge devices, producing data for an 800 x 800 pixel logarithmic heatmap at an average of 2.68 FPS. This study validates the feasibility of decentralized vision-based PET analysis for intelligent transportation systems, offering a replicable methodology for high-resolution, real-time, and scalable intersection safety evaluation.
Authors: Xianglong Shi, Silin Cheng, Sirui Zhao, Yunhan Jiang, Enhong Chen, Yang Liu, Sebastien Ourselin
Abstract: Existing Weakly-Supervised Referring Expression Comprehension (WREC) methods, while effective, are fundamentally limited by a one-to-one mapping assumption, hindering their ability to handle expressions corresponding to zero or multiple targets in realistic scenarios. To bridge this gap, we introduce the Weakly-Supervised Generalized Referring Expression Comprehension task (WGREC), a more practical paradigm that handles expressions with variable numbers of referents. However, extending WREC to WGREC presents two fundamental challenges: supervisory signal ambiguity, where weak image-level supervision is insufficient for training a model to infer the correct number and identity of referents, and semantic representation collapse, where standard Euclidean similarity forces hierarchically-related concepts into non-discriminative clusters, blurring categorical boundaries. To tackle these challenges, we propose a novel WGREC framework named Linguistic Instance-Split Hyperbolic-Euclidean (LIHE), which operates in two stages. The first stage, Referential Decoupling, predicts the number of target objects and decomposes the complex expression into simpler sub-expressions. The second stage, Referent Grounding, then localizes these sub-expressions using HEMix, our innovative hybrid similarity module that synergistically combines the precise alignment capabilities of Euclidean proximity with the hierarchical modeling strengths of hyperbolic geometry. This hybrid approach effectively prevents semantic collapse while preserving fine-grained distinctions between related concepts. Extensive experiments demonstrate LIHE establishes the first effective weakly supervised WGREC baseline on gRefCOCO and Ref-ZOM, while HEMix achieves consistent improvements on standard REC benchmarks, improving IoU@0.5 by up to 2.5\%. The code is available at https://anonymous.4open.science/r/LIHE.
Authors: Jose Reinaldo Cunha Santos A V Silva Neto, Hodaka Kawachi, Yasushi Yagi, Tomoya Nakamura
Abstract: State-of-the-art photorealistic reconstructions for lensless cameras often rely on paired lensless-lensed supervision, which can bias models due to lens-lensless domain mismatch. To avoid this, ground-truth-free diffusion priors are attractive; however, generic formulations tuned for conventional inverse problems often break under the noisy, highly multiplexed, and ill-posed lensless deconvolution setting. We observe that methods which separate range-space enforcement from null-space diffusion-prior updates yield stable, realistic reconstructions. Building on this, we introduce Null-Space Diffusion Distillation (NSDD): a single-pass student that distills the null-space component of an iterative DDNM+ solver, conditioned on the lensless measurement and on a range-space anchor. NSDD preserves measurement consistency and achieves photorealistic results without paired supervision at a fraction of the runtime and memory. On Lensless-FFHQ and PhlatCam, NSDD is the second fastest, behind Wiener, and achieves near-teacher perceptual quality (second-best LPIPS, below DDNM+), outperforming DPS and classical convex baselines. These results suggest a practical path toward fast, ground-truth-free, photorealistic lensless imaging.
Authors: Rulin Zhou, Wenlong He, An Wang, Jianhang Zhang, Xuanhui Zeng, Xi Zhang, Chaowei Zhu, Haijun Hu, Hongliang Ren
Abstract: Accurate point tracking in surgical environments remains challenging due to complex visual conditions, including smoke occlusion, specular reflections, and tissue deformation. While existing surgical tracking datasets provide coordinate information, they lack the semantic context necessary to understand tracking failure mechanisms. We introduce VL-SurgPT, the first large-scale multimodal dataset that bridges visual tracking with textual descriptions of point status in surgical scenes. The dataset comprises 908 in vivo video clips, including 754 for tissue tracking (17,171 annotated points across five challenging scenarios) and 154 for instrument tracking (covering seven instrument types with detailed keypoint annotations). We establish comprehensive benchmarks using eight state-of-the-art tracking methods and propose TG-SurgPT, a text-guided tracking approach that leverages semantic descriptions to improve robustness in visually challenging conditions. Experimental results demonstrate that incorporating point status information significantly improves tracking accuracy and reliability, particularly in adverse visual scenarios where conventional vision-only methods struggle. By bridging visual and linguistic modalities, VL-SurgPT enables the development of context-aware tracking systems crucial for advancing computer-assisted surgery applications that can maintain performance even under challenging intraoperative conditions.
Authors: Jeong Hun Yeo, Sangyun Chung, Sungjune Park, Dae Hoe Kim, Jinyoung Moon, Yong Man Ro
Abstract: Long-video understanding remains a significant challenge for Multimodal Large Language Models (MLLMs) due to inherent token limitations and the complexity of capturing long-term temporal dependencies. Existing methods often fail to capture the global context and complex event relationships necessary for deep video reasoning. To address this, we introduce GCAgent, a novel Global-Context-Aware Agent framework that achieves comprehensive long-video understanding. Our core innovation is the Schematic and Narrative Episodic Memory. This memory structurally models events and their causal and temporal relations into a concise, organized context, fundamentally resolving the long-term dependency problem. Operating in a multi-stage Perception-Action-Reflection cycle, our GCAgent utilizes a Memory Manager to retrieve relevant episodic context for robust, context-aware inference. Extensive experiments confirm that GCAgent significantly enhances long-video understanding, achieving up to 23.5\% accuracy improvement on the Video-MME Long split over a strong MLLM baseline. Furthermore, our framework establishes state-of-the-art performance among comparable 7B-scale MLLMs, achieving 73.4\% accuracy on the Long split and the highest overall average (71.9\%) on the Video-MME benchmark, validating our agent-based reasoning paradigm and structured memory for cognitively-inspired long-video understanding.
Authors: Jun Zhou, Chi Xu, Kaifeng Tang, Yuting Ge, Tingrui Guo, Li Cheng
Abstract: Estimating the 3D poses of hands and objects from a single RGB image is a fundamental yet challenging problem, with broad applications in augmented reality and human-computer interaction. Existing methods largely rely on visual cues alone, often producing results that violate physical constraints such as interpenetration or non-contact. Recent efforts to incorporate physics reasoning typically depend on post-optimization or non-differentiable physics engines, which compromise visual consistency and end-to-end trainability. To overcome these limitations, we propose a novel framework that jointly integrates visual and physical cues for hand-object pose estimation. This integration is achieved through two key ideas: 1) joint visual-physical cue learning: The model is trained to extract 2D visual cues and 3D physical cues, thereby enabling more comprehensive representation learning for hand-object interactions; 2) candidate pose aggregation: A novel refinement process that aggregates multiple diffusion-generated candidate poses by leveraging both visual and physical predictions, yielding a final estimate that is visually consistent and physically plausible. Extensive experiments demonstrate that our method significantly outperforms existing state-of-the-art approaches in both pose accuracy and physical plausibility.
Authors: Guotao Liang, Baoquan Zhang, Zhiyuan Wen, Zihao Han, Yunming Ye
Abstract: Masked image generation (MIG) has demonstrated remarkable efficiency and high-fidelity images by enabling parallel token prediction. Existing methods typically rely solely on the model itself to learn semantic dependencies among visual token sequences. However, directly learning such semantic dependencies from data is challenging because the individual tokens lack clear semantic meanings, and these sequences are usually long. To address this limitation, we propose a novel Knowledge-Augmented Masked Image Generation framework, named KA-MIG, which introduces explicit knowledge of token-level semantic dependencies (\emph{i.e.}, extracted from the training data) as priors to learn richer representations for improving performance. In particular, we explore and identify three types of advantageous token knowledge graphs, including two positive and one negative graphs (\emph{i.e.}, the co-occurrence graph, the semantic similarity graph, and the position-token incompatibility graph). Based on three prior knowledge graphs, we design a graph-aware encoder to learn token and position-aware representations. After that, a lightweight fusion mechanism is introduced to integrate these enriched representations into the existing MIG methods. Resorting to such prior knowledge, our method effectively enhances the model's ability to capture semantic dependencies, leading to improved generation quality. Experimental results demonstrate that our method improves upon existing MIG for class-conditional image generation on ImageNet.
Authors: Xiaohao Liu, Xiaobo Xia, Jiaheng Wei, Shuo Yang, Xiu Su, See-Kiong Ng, Tat-Seng Chua
Abstract: Multimodal representation learning harmonizes distinct modalities by aligning them into a unified latent space. Recent research generalizes traditional cross-modal alignment to produce enhanced multimodal synergy but requires all modalities to be present for a common instance, making it challenging to utilize prevalent datasets with missing modalities. We provide theoretical insights into this issue from an anchor shift perspective. Observed modalities are aligned with a local anchor that deviates from the optimal one when all modalities are present, resulting in an inevitable shift. To address this, we propose CalMRL for multimodal representation learning to calibrate incomplete alignments caused by missing modalities. Specifically, CalMRL leverages the priors and the inherent connections among modalities to model the imputation for the missing ones at the representation level. To resolve the optimization dilemma, we employ a bi-step learning method with the closed-form solution of the posterior distribution of shared latents. We validate its mitigation of anchor shift and convergence with theoretical guidance. By equipping the calibrated alignment with the existing advanced method, we offer new flexibility to absorb data with missing modalities, which is originally unattainable. Extensive experiments and comprehensive analyses demonstrate the superiority of CalMRL. Our code, model checkpoints, and evaluation raw data will be publicly available.
Authors: Xinyuan Hu, Changyue Shi, Chuxiao Yang, Minghao Chen, Jiajun Ding, Tao Wei, Chen Wei, Zhou Yu, Min Tan
Abstract: Feed-forward 3D reconstruction from sparse, low-resolution (LR) images is a crucial capability for real-world applications, such as autonomous driving and embodied AI. However, existing methods often fail to recover fine texture details. This limitation stems from the inherent lack of high-frequency information in LR inputs. To address this, we propose \textbf{SRSplat}, a feed-forward framework that reconstructs high-resolution 3D scenes from only a few LR views. Our main insight is to compensate for the deficiency of texture information by jointly leveraging external high-quality reference images and internal texture cues. We first construct a scene-specific reference gallery, generated for each scene using Multimodal Large Language Models (MLLMs) and diffusion models. To integrate this external information, we introduce the \textit{Reference-Guided Feature Enhancement (RGFE)} module, which aligns and fuses features from the LR input images and their reference twin image. Subsequently, we train a decoder to predict the Gaussian primitives using the multi-view fused feature obtained from \textit{RGFE}. To further refine predicted Gaussian primitives, we introduce \textit{Texture-Aware Density Control (TADC)}, which adaptively adjusts Gaussian density based on the internal texture richness of the LR inputs. Extensive experiments demonstrate that our SRSplat outperforms existing methods on various datasets, including RealEstate10K, ACID, and DTU, and exhibits strong cross-dataset and cross-resolution generalization capabilities.
Authors: Cheng-Chang Tsai, Kai-Wen Cheng, Chun-Shien Lu
Abstract: Federated learning (FL) has shown success in collaboratively training a model among decentralized data resources without directly sharing privacy-sensitive training data. Despite recent advances, non-IID (non-independent and identically distributed) data poses an inevitable challenge that hinders the use of FL. In this work, we address the issue of non-IID histopathological images with feature distribution shifts from an intuitive perspective that has only received limited attention. Specifically, we address this issue from the perspective of data distribution by solely adjusting the data distributions of all clients. Building on the success of diffusion models in fitting data distributions and leveraging stain separation to extract the pivotal features that are closely related to the non-IID properties of histopathological images, we propose a Federated Stain Distribution Alignment (FedSDA) method. FedSDA aligns the stain distribution of each client with a target distribution in an FL framework to mitigate distribution shifts among clients. Furthermore, considering that training diffusion models on raw data in FL has been shown to be susceptible to privacy leakage risks, we circumvent this problem while still effectively achieving alignment. Extensive experimental results show that FedSDA is not only effective in improving baselines that focus on mitigating disparities across clients' model updates but also outperforms baselines that address the non-IID data issues from the perspective of data distribution. We show that FedSDA provides valuable and practical insights for the computational pathology community.
Authors: Huimin Cheng, Xiaowei Yu, Shushan Wu, Luyang Fang, Chao Cao, Jing Zhang, Tianming Liu, Dajiang Zhu, Wenxuan Zhong, Ping Ma
Abstract: Medical images exhibit latent anatomical groupings, such as organs, tissues, and pathological regions, that standard Vision Transformers (ViTs) fail to exploit. While recent work like SBM-Transformer attempts to incorporate such structures through stochastic binary masking, they suffer from non-differentiability, training instability, and the inability to model complex community structure. We present DCMM-Transformer, a novel ViT architecture for medical image analysis that incorporates a Degree-Corrected Mixed-Membership (DCMM) model as an additive bias in self-attention. Unlike prior approaches that rely on multiplicative masking and binary sampling, our method introduces community structure and degree heterogeneity in a fully differentiable and interpretable manner. Comprehensive experiments across diverse medical imaging datasets, including brain, chest, breast, and ocular modalities, demonstrate the superior performance and generalizability of the proposed approach. Furthermore, the learned group structure and structured attention modulation substantially enhance interpretability by yielding attention maps that are anatomically meaningful and semantically coherent.
Authors: Saksham Kumar, Ashish Singh, Srinivasarao Thota, Sunil Kumar Singh, Chandan Kumar
Abstract: Deepfakes are major threats to the integrity of digital media. We propose DeiTFake, a DeiT-based transformer and a novel two-stage progressive training strategy with increasing augmentation complexity. The approach applies an initial transfer-learning phase with standard augmentations followed by a fine-tuning phase using advanced affine and deepfake-specific augmentations. DeiT's knowledge distillation model captures subtle manipulation artifacts, increasing robustness of the detection model. Trained on the OpenForensics dataset (190,335 images), DeiTFake achieves 98.71\% accuracy after stage one and 99.22\% accuracy with an AUROC of 0.9997, after stage two, outperforming the latest OpenForensics baselines. We analyze augmentation impact and training schedules, and provide practical benchmarks for facial deepfake detection.
Authors: Cuiqun Chen, Qi Chen, Bin Yang, Xingyi Zhang
Abstract: Cross-view geo-localization (CVGL) matches query images ($\textit{e.g.}$, drone) to geographically corresponding opposite-view imagery ($\textit{e.g.}$, satellite). While supervised methods achieve strong performance, their reliance on extensive pairwise annotations limits scalability. Unsupervised alternatives avoid annotation costs but suffer from noisy pseudo-labels due to intrinsic cross-view domain gaps. To address these limitations, we propose $\textit{UniABG}$, a novel dual-stage unsupervised cross-view geo-localization framework integrating adversarial view bridging with graph-based correspondence calibration. Our approach first employs View-Aware Adversarial Bridging (VAAB) to model view-invariant features and enhance pseudo-label robustness. Subsequently, Heterogeneous Graph Filtering Calibration (HGFC) refines cross-view associations by constructing dual inter-view structure graphs, achieving reliable view correspondence. Extensive experiments demonstrate state-of-the-art unsupervised performance, showing that UniABG improves Satellite $\rightarrow$ Drone AP by +10.63\% on University-1652 and +16.73\% on SUES-200, even surpassing supervised baselines. The source code is available at https://github.com/chenqi142/UniABG
Authors: Sijie Wang, Qiang Wang, Shaohuai Shi
Abstract: Video generation has been advancing rapidly, and diffusion transformer (DiT) based models have demonstrated remark- able capabilities. However, their practical deployment is of- ten hindered by slow inference speeds and high memory con- sumption. In this paper, we propose a novel pipelining frame- work named PipeDiT to accelerate video generation, which is equipped with three main innovations. First, we design a pipelining algorithm (PipeSP) for sequence parallelism (SP) to enable the computation of latent generation and commu- nication among multiple GPUs to be pipelined, thus reduc- ing inference latency. Second, we propose DeDiVAE to de- couple the diffusion module and the variational autoencoder (VAE) module into two GPU groups, whose executions can also be pipelined to reduce memory consumption and infer- ence latency. Third, to better utilize the GPU resources in the VAE group, we propose an attention co-processing (Aco) method to further reduce the overall video generation latency. We integrate our PipeDiT into both OpenSoraPlan and Hun- yuanVideo, two state-of-the-art open-source video generation frameworks, and conduct extensive experiments on two 8- GPU systems. Experimental results show that, under many common resolution and timestep configurations, our PipeDiT achieves 1.06x to 4.02x speedups over OpenSoraPlan and HunyuanVideo.
Authors: Zhichen Lai, Hua Lu, Huan Li, Jialiang Li, Christian S. Jensen
Abstract: Trajectory similarity computation is fundamental functionality that is used for, e.g., clustering, prediction, and anomaly detection. However, existing learning-based methods exhibit three key limitations: (1) insufficient modeling of trajectory semantics and hierarchy, lacking both movement dynamics extraction and multi-scale structural representation; (2) high computational costs due to point-wise encoding; and (3) use of physically implausible augmentations that distort trajectory semantics. To address these issues, we propose MovSemCL, a movement-semantics contrastive learning framework for trajectory similarity computation. MovSemCL first transforms raw GPS trajectories into movement-semantics features and then segments them into patches. Next, MovSemCL employs intra- and inter-patch attentions to encode local as well as global trajectory patterns, enabling efficient hierarchical representation and reducing computational costs. Moreover, MovSemCL includes a curvature-guided augmentation strategy that preserves informative segments (e.g., turns and intersections) and masks redundant ones, generating physically plausible augmented views. Experiments on real-world datasets show that MovSemCL is capable of outperforming state-of-the-art methods, achieving mean ranks close to the ideal value of 1 at similarity search tasks and improvements by up to 20.3% at heuristic approximation, while reducing inference latency by up to 43.4%.
Authors: Jialang Lu, Shuning Sun, Pu Wang, Chen Wu, Feng Gao, Lina Gong, Dianjie Lu, Guijuan Zhang, Zhuoran Zheng
Abstract: Purple fringing, a persistent artifact caused by Longitudinal Chromatic Aberration (LCA) in camera lenses, has long degraded the clarity and realism of digital imaging. Traditional solutions rely on complex and expensive apochromatic (APO) lens hardware and the extraction of handcrafted features, ignoring the data-driven approach. To fill this gap, we introduce DCA-LUT, the first deep learning framework for purple fringing removal. Inspired by the physical root of the problem, the spatial misalignment of RGB color channels due to lens dispersion, we introduce a novel Chromatic-Aware Coordinate Transformation (CA-CT) module, learning an image-adaptive color space to decouple and isolate fringing into a dedicated dimension. This targeted separation allows the network to learn a precise ``purple fringe channel", which then guides the accurate restoration of the luminance channel. The final color correction is performed by a learned 5D Look-Up Table (5D LUT), enabling efficient and powerful% non-linear color mapping. To enable robust training and fair evaluation, we constructed a large-scale synthetic purple fringing dataset (PF-Synth). Extensive experiments in synthetic and real-world datasets demonstrate that our method achieves state-of-the-art performance in purple fringing removal.
Authors: Dengming Zhang, Weitao You, Jingxiong Li, Weishen Lin, Wenda Shi, Xue Zhao, Heda Zuo, Junxian Wu, Lingyun Sun
Abstract: Emotion understanding is critical for making Large Language Models (LLMs) more general, reliable, and aligned with humans. Art conveys emotion through the joint design of visual and auditory elements, yet most prior work is human-centered or single-modality, overlooking the emotion intentionally expressed by the artwork. Meanwhile, current Audio-Visual Language Models (AVLMs) typically require large-scale audio pretraining to endow Visual Language Models (VLMs) with hearing, which limits scalability. We present Vision Anchored Audio-Visual Emotion LLM (VAEmotionLLM), a two-stage framework that teaches a VLM to hear by seeing with limited audio pretraining and to understand emotion across modalities. In Stage 1, Vision-Guided Audio Alignment (VG-Align) distills the frozen visual pathway into a new audio pathway by aligning next-token distributions of the shared LLM on synchronized audio-video clips, enabling hearing without a large audio dataset. In Stage 2, a lightweight Cross-Modal Emotion Adapter (EmoAdapter), composed of the Emotion Enhancer and the Emotion Supervisor, injects emotion-sensitive residuals and applies emotion supervision to enhance cross-modal emotion understanding. We also construct ArtEmoBenchmark, an art-centric emotion benchmark that evaluates content and emotion understanding under audio-only, visual-only, and audio-visual inputs. VAEmotionLLM achieves state-of-the-art results on ArtEmoBenchmark, outperforming audio-only, visual-only, and audio-visual baselines. Ablations show that the proposed components are complementary.
Authors: Hongxuan Li (College of Intelligence and Computing, Tianjin University), Wencheng Zhu (College of Intelligence and Computing, Tianjin University, Haihe Laboratory of Information Technology Application Innovation), Huiying Xu (School of Computer Science and Technology, Zhejiang Normal University), Xinzhong Zhu (School of Computer Science and Technology, Zhejiang Normal University), Pengfei Zhu (College of Intelligence and Computing, Tianjin University)
Abstract: Vector quantization has emerged as a powerful tool in large-scale multimodal models, unifying heterogeneous representations through discrete token encoding. However, its effectiveness hinges on robust codebook design. Current prototype-based approaches relying on trainable vectors or clustered centroids fall short in representativeness and interpretability, even as multimodal alignment demonstrates its promise in vision-language models. To address these limitations, we propose a simple multimodal prompting-driven quantization framework for point cloud analysis. Our methodology is built upon two core insights: 1) Text embeddings from pre-trained models inherently encode visual semantics through many-to-one contrastive alignment, naturally serving as robust prototype priors; and 2) Multimodal prompts enable adaptive refinement of these prototypes, effectively mitigating vision-language semantic gaps. The framework introduces a dual-constrained quantization space, enforced by compactness and separation regularization, which seamlessly integrates visual and prototype features, resulting in hybrid representations that jointly encode geometric and semantic information. Furthermore, we employ Gumbel-Softmax relaxation to achieve differentiable discretization while maintaining quantization sparsity. Extensive experiments on the ModelNet40 and ScanObjectNN datasets clearly demonstrate the superior effectiveness of the proposed method.
Authors: Lokender Singh, Saksham Kumar, Chandan Kumar
Abstract: Multilabel image categorization has drawn interest recently because of its numerous computer vision applications. The proposed work introduces a novel method for classifying multilabel images using the COCO-2014 dataset and a modified ResNet-101 architecture. By simulating label dependencies and uncertainties, the approach uses probabilistic reasoning to improve prediction accuracy. Extensive tests show that the model outperforms earlier techniques and approaches to state-of-the-art outcomes in multilabel categorization. The work also thoroughly assesses the model's performance using metrics like precision-recall score and achieves 0.794 mAP on COCO-2014, outperforming ResNet-SRN (0.771) and Vision Transformer baselines (0.785). The novelty of the work lies in integrating probabilistic reasoning into deep learning models to effectively address the challenges presented by multilabel scenarios.
Authors: Ji-Ping Jin, Chen-Bin Feng, Rui Fan, Chi-Man Vong
Abstract: Image stitching often faces challenges due to varying capture angles, positional differences, and object movements, leading to misalignments and visual discrepancies. Traditional seam carving methods neglect semantic information, causing disruptions in foreground continuity. We introduce SemanticStitch, a deep learning-based framework that incorporates semantic priors of foreground objects to preserve their integrity and enhance visual coherence. Our approach includes a novel loss function that emphasizes the semantic integrity of salient objects, significantly improving stitching quality. We also present two specialized real-world datasets to evaluate our method's effectiveness. Experimental results demonstrate substantial improvements over traditional techniques, providing robust support for practical applications.
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: Shuhan Ye, Yi Yu, Qixin Zhang, Chenqi Kong, Qiangqiang Wu, Kun Wang, Xudong Jiang
Abstract: Event cameras sense brightness changes and output binary asynchronous event streams, attracting increasing attention. Their bio-inspired dynamics align well with spiking neural networks (SNNs), offering a promising energy-efficient alternative to conventional vision systems. However, SNNs remain costly to train due to temporal coding, which limits their practical deployment. To alleviate the high training cost of SNNs, we introduce \textbf{PACE} (Phase-Aligned Condensation for Events), the first dataset distillation framework to SNNs and event-based vision. PACE distills a large training dataset into a compact synthetic one that enables fast SNN training, which is achieved by two core modules: \textbf{ST-DSM} and \textbf{PEQ-N}. ST-DSM uses residual membrane potentials to densify spike-based features (SDR) and to perform fine-grained spatiotemporal matching of amplitude and phase (ST-SM), while PEQ-N provides a plug-and-play straight through probabilistic integer quantizer compatible with standard event-frame pipelines. Across DVS-Gesture, CIFAR10-DVS, and N-MNIST datasets, PACE outperforms existing coreset selection and dataset distillation baselines, with particularly strong gains on dynamic event streams and at low or moderate IPC. Specifically, on N-MNIST, it achieves \(84.4\%\) accuracy, about \(85\%\) of the full training set performance, while reducing training time by more than \(50\times\) and storage cost by \(6000\times\), yielding compact surrogates that enable minute-scale SNN training and efficient edge deployment.
Authors: Shuhan Ye, Yi Yu, Qixin Zhang, Chenqi Kong, Qiangqiang Wu, Xudong Jiang, Dacheng Tao
Abstract: Brain-inspired Spiking neural networks (SNNs) promise energy-efficient intelligence via event-driven, sparse computation, but deeper architectures inflate parameters and computational cost, hindering their edge deployment. Recent progress in SNN pruning helps alleviate this burden, yet existing efforts fall into only two families: \emph{unstructured} pruning, which attains high sparsity but is difficult to accelerate on general hardware, and \emph{structured} pruning, which eases deployment but lack flexibility and often degrades accuracy at matched sparsity. In this work, we introduce \textbf{SpikeNM}, the first SNN-oriented \emph{semi-structured} \(N{:}M\) pruning framework that learns sparse SNNs \emph{from scratch}, enforcing \emph{at most \(N\)} non-zeros per \(M\)-weight block. To avoid the combinatorial space complexity \(\sum_{k=1}^{N}\binom{M}{k}\) growing exponentially with \(M\), SpikeNM adopts an \(M\)-way basis-logit parameterization with a differentiable top-\(k\) sampler, \emph{linearizing} per-block complexity to \(\mathcal O(M)\) and enabling more aggressive sparsification. Further inspired by neuroscience, we propose \emph{eligibility-inspired distillation} (EID), which converts temporally accumulated credits into block-wise soft targets to align mask probabilities with spiking dynamics, reducing sampling variance and stabilizing search under high sparsity. Experiments show that at \(2{:}4\) sparsity, SpikeNM maintains and even with gains across main-stream datasets, while yielding hardware-amenable patterns that complement intrinsic spike sparsity.
Authors: Xianhao Zhou, Jianghao Wu, Ku Zhao, Jinlong He, Huangxuan Zhao, Lei Chen, Shaoting Zhang, Guotai Wang
Abstract: Generating synthetic CT images from CBCT or MRI has a potential for efficient radiation dose planning and adaptive radiotherapy. However, existing CNN-based models lack global semantic understanding, while Transformers often overfit small medical datasets due to high model capacity and weak inductive bias. To address these limitations, we propose a DINOv3-Guided Cross Fusion (DGCF) framework that integrates a frozen self-supervised DINOv3 Transformer with a trainable CNN encoder-decoder. It hierarchically fuses global representation of Transformer and local features of CNN via a learnable cross fusion module, achieving balanced local appearance and contextual representation. Furthermore, we introduce a Multi-Level DINOv3 Perceptual (MLDP) loss that encourages semantic similarity between synthetic CT and the ground truth in DINOv3's feature space. Experiments on the SynthRAD2023 pelvic dataset demonstrate that DGCF achieved state-of-the-art performance in terms of MS-SSIM, PSNR and segmentation-based metrics on both MRI$\rightarrow$CT and CBCT$\rightarrow$CT translation tasks. To the best of our knowledge, this is the first work to employ DINOv3 representations for medical image translation, highlighting the potential of self-supervised Transformer guidance for semantic-aware CT synthesis. The code is available at https://github.com/HiLab-git/DGCF.
Authors: Tianle Cheng, Zeyan Zhang, Kaifeng Gao, Jun Xiao
Abstract: Recent advancements in diffusion-based video generation have produced impressive and high-fidelity short videos. To extend these successes to generate coherent long videos, most video diffusion models (VDMs) generate videos in an autoregressive manner, i.e., generating subsequent frames conditioned on previous ones. There are generally two primary paradigms: chunk-based extension and stream denoising. The former directly concatenates previous clean frames as conditioning, suffering from denoising latency and error accumulation. The latter maintains the denoising sequence with monotonically increasing noise levels. In each denoising iteration, one clean frame is produced while a new pure noise is simultaneously appended, enabling live-stream sampling. However, it struggles with fragile consistency and poor motion dynamics. In this paper, we propose Adaptive Begin-of-Video Tokens (ada-BOV) for autoregressive VDMs. The BOV tokens are special learnable embeddings on VDMs. They adaptively absorb denoised preceding frames via an adaptive-layer-norm-like modulation. This design preserves the global consistency while allowing for flexible conditioning in dynamic scenarios. To ensure the quality of local dynamics essential in modulating BOV tokens, we further propose a refinement strategy for stream denoising. It decouples the sampling trajectory length from the attention window size constraint, leading to improved local guidance and overall imaging quality. We also propose a disturbance-augmented training noise schedule, which balances the convergence speed with model robustness for the stream denoising. Extensive experiments demonstrate that our method achieves compelling qualitative and quantitative results across multiple metrics.
Authors: Yannan Chen, Ruoyu Chen, Bin Zeng, Wei Wang, Shiming Liu, Qunli Zhang, Zheng Hu, Laiyuan Wang, Yaowei Wang, Xiaochun Cao
Abstract: In current visual model training, models often rely on only limited sufficient causes for their predictions, which makes them sensitive to distribution shifts or the absence of key features. Attribution methods can accurately identify a model's critical regions. However, masking these areas to create counterfactuals often causes the model to misclassify the target, while humans can still easily recognize it. This divergence highlights that the model's learned dependencies may not be sufficiently causal. To address this issue, we propose Subset-Selected Counterfactual Augmentation (SS-CA), which integrates counterfactual explanations directly into the training process for targeted intervention. Building on the subset-selection-based LIMA attribution method, we develop Counterfactual LIMA to identify minimal spatial region sets whose removal can selectively alter model predictions. Leveraging these attributions, we introduce a data augmentation strategy that replaces the identified regions with natural background, and we train the model jointly on both augmented and original samples to mitigate incomplete causal learning. Extensive experiments across multiple ImageNet variants show that SS-CA improves generalization on in-distribution (ID) test data and achieves superior performance on out-of-distribution (OOD) benchmarks such as ImageNet-R and ImageNet-S. Under perturbations including noise, models trained with SS-CA also exhibit enhanced generalization, demonstrating that our approach effectively uses interpretability insights to correct model deficiencies and improve both performance and robustness.
Authors: Sayad Ibna Azad, Md. Atiqur Rahman
Abstract: We introduce BdSL-SPOTER, a pose-based transformer framework for accurate and efficient recognition of Bengali Sign Language (BdSL). BdSL-SPOTER extends the SPOTER paradigm with cultural specific preprocessing and a compact four-layer transformer encoder featuring optimized learnable positional encodings, while employing curriculum learning to enhance generalization on limited data and accelerate convergence. On the BdSLW60 benchmark, it achieves 97.92% Top-1 validation accuracy, representing a 22.82% improvement over the Bi-LSTM baseline, all while keeping computational costs low. With its reduced number of parameters, lower FLOPs, and higher FPS, BdSL-SPOTER provides a practical framework for real-world accessibility applications and serves as a scalable model for other low-resource regional sign languages.
Authors: Tammy Glazer, Gilles Q. Hacheme, Akram Zaytar, Luana Marotti, Amy Michaels, Girmaw Abebe Tadesse, Kevin White, Rahul Dodhia, Andrew Zolli, Inbal Becker-Reshef, Juan M. Lavista Ferres, Caleb Robinson
Abstract: We present TEMPO, a global, temporally resolved dataset of building density and height derived from high-resolution satellite imagery using deep learning models. We pair building footprint and height data from existing datasets with quarterly PlanetScope basemap satellite images to train a multi-task deep learning model that predicts building density and building height at a 37.6-meter per pixel resolution. We apply this model to global PlanetScope basemaps from Q1 2018 through Q2 2025 to create global, temporal maps of building density and height. We validate these maps by comparing against existing building footprint datasets. Our estimates achieve an F1 score between 85% and 88% on different hand-labeled subsets, and are temporally stable, with a 0.96 five-year trend-consistency score. TEMPO captures quarterly changes in built settlements at a fraction of the computational cost of comparable approaches, unlocking large-scale monitoring of development patterns and climate impacts essential for global resilience and adaptation efforts.
Authors: Tianxiang Zhang, Peipeng Yu, Zhihua Xia, Longchen Dai, Xiaoyu Zhou, Hui Gao
Abstract: The proliferation of sophisticated deepfakes poses significant threats to information integrity. While DINOv2 shows promise for detection, existing fine-tuning approaches treat it as generic binary classification, overlooking distinct artifacts inherent to different deepfake methods. To address this, we propose a DeepFake Fine-Grained Adapter (DFF-Adapter) for DINOv2. Our method incorporates lightweight multi-head LoRA modules into every transformer block, enabling efficient backbone adaptation. DFF-Adapter simultaneously addresses authenticity detection and fine-grained manipulation type classification, where classifying forgery methods enhances artifact sensitivity. We introduce a shared branch propagating fine-grained manipulation cues to the authenticity head. This enables multi-task cooperative optimization, explicitly enhancing authenticity discrimination with manipulation-specific knowledge. Utilizing only 3.5M trainable parameters, our parameter-efficient approach achieves detection accuracy comparable to or even surpassing that of current complex state-of-the-art methods.
Authors: Qinyue Tong, Ziqian Lu, Jun Liu, Rui Zuo, Zheming Lu
Abstract: Despite the progress in medical image segmentation, most existing methods remain task-specific and lack interactivity. Although recent text-prompt-based segmentation approaches enhance user-driven and reasoning-based segmentation, they remain confined to single-round dialogues and fail to perform multi-round reasoning. In this work, we introduce Multi-Round Entity-Level Medical Reasoning Segmentation (MEMR-Seg), a new task that requires generating segmentation masks through multi-round queries with entity-level reasoning. To support this task, we construct MR-MedSeg, a large-scale dataset of 177K multi-round medical segmentation dialogues, featuring entity-based reasoning across rounds. Furthermore, we propose MediRound, an effective baseline model designed for multi-round medical reasoning segmentation. To mitigate the inherent error propagation in the chain-like pipeline of multi-round segmentation, we introduce a lightweight yet effective Judgment & Correction Mechanism during model inference. Experimental results demonstrate that our method effectively addresses the MEMR-Seg task and outperforms conventional medical referring segmentation methods.
Authors: Ruiqi Cheng, Huijun Di, Jian Li, Feng Liu, Wei Liang
Abstract: Accurate 3D scene motion perception significantly enhances the safety and reliability of an autonomous driving system. Benefiting from its all-weather operational capability and unique perceptual properties, 4D mmWave radar has emerged as an essential component in advanced autonomous driving. However, sparse and noisy radar points often lead to imprecise motion perception, leaving autonomous vehicles with limited sensing capabilities when optical sensors degrade under adverse weather conditions. In this paper, we propose RadarMP, a novel method for precise 3D scene motion perception using low-level radar echo signals from two consecutive frames. Unlike existing methods that separate radar target detection and motion estimation, RadarMP jointly models both tasks in a unified architecture, enabling consistent radar point cloud generation and pointwise 3D scene flow prediction. Tailored to radar characteristics, we design specialized self-supervised loss functions guided by Doppler shifts and echo intensity, effectively supervising spatial and motion consistency without explicit annotations. Extensive experiments on the public dataset demonstrate that RadarMP achieves reliable motion perception across diverse weather and illumination conditions, outperforming radar-based decoupled motion perception pipelines and enhancing perception capabilities for full-scenario autonomous driving systems.
Authors: Quanxing Xu, Ling Zhou, Feifei Zhang, Jinyu Tian, Rubing Huang
Abstract: Large Language Models (LLMs) have become a crucial tool in Visual Question Answering (VQA) for handling knowledge-intensive questions in few-shot or zero-shot scenarios. However, their reliance on massive training datasets often causes them to inherit language biases during the acquisition of knowledge. This limitation imposes two key constraints on existing methods: (1) LLM predictions become less reliable due to bias exploitation, and (2) despite strong knowledge reasoning capabilities, LLMs still struggle with out-of-distribution (OOD) generalization. To address these issues, we propose Object Attribute Description Promoter (OAD-Promoter), a novel approach for enhancing LLM-based VQA by mitigating language bias and improving domain-shift robustness. OAD-Promoter comprises three components: the Object-concentrated Example Generation (OEG) module, the Memory Knowledge Assistance (MKA) module, and the OAD Prompt. The OEG module generates global captions and object-concentrated samples, jointly enhancing visual information input to the LLM and mitigating bias through complementary global and regional visual cues. The MKA module assists the LLM in handling OOD samples by retrieving relevant knowledge from stored examples to support questions from unseen domains. Finally, the OAD Prompt integrates the outputs of the preceding modules to optimize LLM inference. Experiments demonstrate that OAD-Promoter significantly improves the performance of LLM-based VQA methods in few-shot or zero-shot settings, achieving new state-of-the-art results.
Authors: Karol C. Jurzec, Tomasz Szydlo, Maciej Wielgosz
Abstract: Spiking neural networks (SNNs) communicate via discrete spikes in time rather than continuous activations. Their event-driven nature offers advantages for temporal processing and energy efficiency on resource-constrained hardware, but training and deployment remain challenging. We present a lightweight C-based runtime for SNN inference on edge devices and optimizations that reduce latency and memory without sacrificing accuracy. Trained models exported from SNNTorch are translated to a compact C representation; static, cache-friendly data layouts and preallocation avoid interpreter and allocation overheads. We further exploit sparse spiking activity to prune inactive neurons and synapses, shrinking computation in upstream convolutional layers. Experiments on N-MNIST and ST-MNIST show functional parity with the Python baseline while achieving ~10 speedups on desktop CPU and additional gains with pruning, together with large memory reductions that enable microcontroller deployment (Arduino Portenta H7). Results indicate that SNNs can be executed efficiently on conventional embedded platforms when paired with an optimized runtime and spike-driven model compression. Code: https://github.com/karol-jurzec/snn-generator/
Authors: Seokwon Song, Minsu Park, Gunhee Kim
Abstract: Source attribution aims to enhance the reliability of AI-generated answers by including references for each statement, helping users validate the provided answers. However, existing work has primarily focused on text-only scenario and largely overlooked the role of multimodality. We introduce MAVIS, the first benchmark designed to evaluate multimodal source attribution systems that understand user intent behind visual questions, retrieve multimodal evidence, and generate long-form answers with citations. Our dataset comprises 157K visual QA instances, where each answer is annotated with fact-level citations referring to multimodal documents. We develop fine-grained automatic metrics along three dimensions of informativeness, groundedness, and fluency, and demonstrate their strong correlation with human judgments. Our key findings are threefold: (1) LVLMs with multimodal RAG generate more informative and fluent answers than unimodal RAG, but they exhibit weaker groundedness for image documents than for text documents, a gap amplified in multimodal settings. (2) Given the same multimodal documents, there is a trade-off between informativeness and groundedness across different prompting methods. (3) Our proposed method highlights mitigating contextual bias in interpreting image documents as a crucial direction for future research. The dataset and experimental code are available at https://github.com/seokwon99/MAVIS
Authors: Yuqi Xie, Shuhan Ye, Yi Yu, Chong Wang, Qixin Zhang, Jiazhen Xu, Le Shen, Yuanbin Qian, Jiangbo Qian, Guoqi Li
Abstract: The integration of event cameras and spiking neural networks (SNNs) promises energy-efficient visual intelligence, yet scarce event data and the sparsity of DVS outputs hinder effective training. Prior knowledge transfers from RGB to DVS often underperform because the distribution gap between modalities is substantial. In this work, we present Time-step Mixup Knowledge Transfer (TMKT), a cross-modal training framework with a probabilistic Time-step Mixup (TSM) strategy. TSM exploits the asynchronous nature of SNNs by interpolating RGB and DVS inputs at various time steps to produce a smooth curriculum within each sequence, which reduces gradient variance and stabilizes optimization with theoretical analysis. To employ auxiliary supervision from TSM, TMKT introduces two lightweight modality-aware objectives, Modality Aware Guidance (MAG) for per-frame source supervision and Mixup Ratio Perception (MRP) for sequence-level mix ratio estimation, which explicitly align temporal features with the mixing schedule. TMKT enables smoother knowledge transfer, helps mitigate modality mismatch during training, and achieves superior performance in spiking image classification tasks. Extensive experiments across diverse benchmarks and multiple SNN backbones, together with ablations, demonstrate the effectiveness of our method.
Authors: Kaixiang Yang, Boyang Shen, Xin Li, Yuchen Dai, Yuxuan Luo, Yueran Ma, Wei Fang, Qiang Li, Zhiwei Wang
Abstract: Text-guided image editing has advanced rapidly with the rise of diffusion models. While flow-based inversion-free methods offer high efficiency by avoiding latent inversion, they often fail to effectively integrate source information, leading to poor background preservation, spatial inconsistencies, and over-editing due to the lack of effective integration of source information. In this paper, we present FIA-Edit, a novel inversion-free framework that achieves high-fidelity and semantically precise edits through a Frequency-Interactive Attention. Specifically, we design two key components: (1) a Frequency Representation Interaction (FRI) module that enhances cross-domain alignment by exchanging frequency components between source and target features within self-attention, and (2) a Feature Injection (FIJ) module that explicitly incorporates source-side queries, keys, values, and text embeddings into the target branch's cross-attention to preserve structure and semantics. Comprehensive and extensive experiments demonstrate that FIA-Edit supports high-fidelity editing at low computational cost (~6s per 512 * 512 image on an RTX 4090) and consistently outperforms existing methods across diverse tasks in visual quality, background fidelity, and controllability. Furthermore, we are the first to extend text-guided image editing to clinical applications. By synthesizing anatomically coherent hemorrhage variations in surgical images, FIA-Edit opens new opportunities for medical data augmentation and delivers significant gains in downstream bleeding classification. Our project is available at: https://github.com/kk42yy/FIA-Edit.
Authors: Shuo Yin, Zhiyuan Yin, Yuqing Hou, Rui Liu, Yong Chen, Dell Zhang
Abstract: Hash center-based deep hashing methods improve upon pairwise or triplet-based approaches by assigning fixed hash centers to each class as learning targets, thereby avoiding the inefficiency of local similarity optimization. However, random center initialization often disregards inter-class semantic relationships. While existing two-stage methods mitigate this by first refining hash centers with semantics and then training the hash function, they introduce additional complexity, computational overhead, and suboptimal performance due to stage-wise discrepancies. To address these limitations, we propose $\textbf{Center-Reassigned Hashing (CRH)}$, an end-to-end framework that $\textbf{dynamically reassigns hash centers}$ from a preset codebook while jointly optimizing the hash function. Unlike previous methods, CRH adapts hash centers to the data distribution $\textbf{without explicit center optimization phases}$, enabling seamless integration of semantic relationships into the learning process. Furthermore, $\textbf{a multi-head mechanism}$ enhances the representational capacity of hash centers, capturing richer semantic structures. Extensive experiments on three benchmarks demonstrate that CRH learns semantically meaningful hash centers and outperforms state-of-the-art deep hashing methods in retrieval tasks.
Authors: Wang Luo, Di Wu, Hengyuan Na, Yinlin Zhu, Miao Hu, Guocong Quan
Abstract: Point cloud completion aims to reconstruct complete 3D shapes from partial observations, which is a challenging problem due to severe occlusions and missing geometry. Despite recent advances in multimodal techniques that leverage complementary RGB images to compensate for missing geometry, most methods still follow a Completion-by-Inpainting paradigm, synthesizing missing structures from fused latent features. We empirically show that this paradigm often results in structural inconsistencies and topological artifacts due to limited geometric and semantic constraints. To address this, we rethink the task and propose a more robust paradigm, termed Completion-by-Correction, which begins with a topologically complete shape prior generated by a pretrained image-to-3D model and performs feature-space correction to align it with the partial observation. This paradigm shifts completion from unconstrained synthesis to guided refinement, enabling structurally consistent and observation-aligned reconstruction. Building upon this paradigm, we introduce PGNet, a multi-stage framework that conducts dual-feature encoding to ground the generative prior, synthesizes a coarse yet structurally aligned scaffold, and progressively refines geometric details via hierarchical correction. Experiments on the ShapeNetViPC dataset demonstrate the superiority of PGNet over state-of-the-art baselines in terms of average Chamfer Distance (-23.5%) and F-score (+7.1%).
Authors: Jinyuan Hu, Jiayou Zhang, Shaobo Cui, Kun Zhang, Guangyi Chen
Abstract: Autoregressive (AR) approaches, which represent images as sequences of discrete tokens from a finite codebook, have achieved remarkable success in image generation. However, the quantization process and the limited codebook size inevitably discard fine-grained information, placing bottlenecks on fidelity. Motivated by this limitation, recent studies have explored autoregressive modeling in continuous latent spaces, which offers higher generation quality. Yet, unlike discrete tokens constrained by a fixed codebook, continuous representations lie in a vast and unstructured space, posing significant challenges for efficient autoregressive modeling. To address these challenges, we introduce MixAR, a novel framework that leverages mixture training paradigms to inject discrete tokens as prior guidance for continuous AR modeling. MixAR is a factorized formulation that leverages discrete tokens as prior guidance for continuous autoregressive prediction. We investigate several discrete-continuous mixture strategies, including self-attention (DC-SA), cross-attention (DC-CA), and a simple approach (DC-Mix) that replaces homogeneous mask tokens with informative discrete counterparts. Moreover, to bridge the gap between ground-truth training tokens and inference tokens produced by the pre-trained AR model, we propose Training-Inference Mixture (TI-Mix) to achieve consistent training and generation distributions. In our experiments, we demonstrate a favorable balance of the DC-Mix strategy between computational efficiency and generation fidelity, and consistent improvement of TI-Mix.
Authors: Abdelrahman Elsayed, Ahmed Jaheen, Mohammad Yaqub
Abstract: Automated brain tumor segmentation in multi-parametric MRI remains challenging in resource-constrained settings where deep 3D networks are computationally prohibitive. We propose MMRINet, a lightweight architecture that replaces quadratic-complexity attention with linear-complexity Mamba state-space models for efficient volumetric context modeling. Novel Dual-Path Feature Refinement (DPFR) modules maximize feature diversity without additional data requirements, while Progressive Feature Aggregation (PFA) enables effective multi-scale fusion. In the BraTS-Lighthouse SSA 2025, our model achieves strong performance with an average Dice score of (0.752) and an average HD95 of (12.23) with only ~2.5M parameters, demonstrating efficient and accurate segmentation suitable for low-resource clinical environments. Our GitHub repository can be accessed here: github.com/BioMedIA-MBZUAI/MMRINet.
Authors: Aditi Bhalla, Christian Hellert, Enkelejda Kasneci
Abstract: Driver distraction remains a leading cause of road traffic accidents, contributing to thousands of fatalities annually across the globe. While deep learning-based driver activity recognition methods have shown promise in detecting such distractions, their effectiveness in real-world deployments is hindered by two critical challenges: variations in camera viewpoints (cross-view) and domain shifts such as change in sensor modality or environment. Existing methods typically address either cross-view generalization or unsupervised domain adaptation in isolation, leaving a gap in the robust and scalable deployment of models across diverse vehicle configurations. In this work, we propose a novel two-phase cross-view, cross-modal unsupervised domain adaptation framework that addresses these challenges jointly on real-time driver monitoring data. In the first phase, we learn view-invariant and action-discriminative features within a single modality using contrastive learning on multi-view data. In the second phase, we perform domain adaptation to a new modality using information bottleneck loss without requiring any labeled data from the new domain. We evaluate our approach using state-of-the art video transformers (Video Swin, MViT) and multi modal driver activity dataset called Drive&Act, demonstrating that our joint framework improves top-1 accuracy on RGB video data by almost 50% compared to a supervised contrastive learning-based cross-view method, and outperforms unsupervised domain adaptation-only methods by up to 5%, using the same video transformer backbone.
Authors: Sujun Sun, Haowen Gu, Cheng Xie, Yanxu Ren, Mingwu Ren, Haofeng Zhang
Abstract: Cross-domain Few-shot Segmentation (CD-FSS) aims to segment novel classes from target domains that are not involved in training and have significantly different data distributions from the source domain, using only a few annotated samples, and recent years have witnessed significant progress on this task. However, existing CD-FSS methods primarily focus on style gaps between source and target domains while ignoring segmentation granularity gaps, resulting in insufficient semantic discriminability for novel classes in target domains. Therefore, we propose a Hierarchical Semantic Learning (HSL) framework to tackle this problem. Specifically, we introduce a Dual Style Randomization (DSR) module and a Hierarchical Semantic Mining (HSM) module to learn hierarchical semantic features, thereby enhancing the model's ability to recognize semantics at varying granularities. DSR simulates target domain data with diverse foreground-background style differences and overall style variations through foreground and global style randomization respectively, while HSM leverages multi-scale superpixels to guide the model to mine intra-class consistency and inter-class distinction at different granularities. Additionally, we also propose a Prototype Confidence-modulated Thresholding (PCMT) module to mitigate segmentation ambiguity when foreground and background are excessively similar. Extensive experiments are conducted on four popular target domain datasets, and the results demonstrate that our method achieves state-of-the-art performance.
Authors: Feng Chen, Yefei He, Shaoxuan He, Yuanyu He, Jing Liu, Lequan Lin, Akide Liu, Zhaoyang Li, Jiyuan Zhang, Zhenbang Sun, Bohan Zhuang, Qi Wu
Abstract: Existing sparse attention methods primarily target inference-time acceleration by selecting critical tokens under predefined sparsity patterns. However, they often fail to bridge the training-inference gap and lack the capacity for fine-grained token selection across multiple dimensions such as queries, key-values (KV), and heads, leading to suboptimal performance and limited acceleration gains. In this paper, we introduce OmniSparse, a training-aware fine-grained sparse attention framework for long-video MLLMs, which operates in both training and inference with dynamic token budget allocation. Specifically, OmniSparse contains three adaptive and complementary mechanisms: (1) query selection via lazy-active classification, retaining active queries that capture broad semantic similarity while discarding most lazy ones that focus on limited local context and exhibit high functional redundancy; (2) KV selection with head-level dynamic budget allocation, where a shared budget is determined based on the flattest head and applied uniformly across all heads to ensure attention recall; and (3) KV cache slimming to reduce head-level redundancy by selectively fetching visual KV cache according to the head-level decoding query pattern. Experimental results show that OmniSparse matches the performance of full attention while achieving up to 2.7x speedup during prefill and 2.4x memory reduction during decoding.
Authors: Zhuojiang Cai, Yiheng Zhang, Meitong Guo, Mingdao Wang, Yuwang Wang
Abstract: Recently, multi-view diffusion-based 3D generation methods have gained significant attention. However, these methods often suffer from shape and texture misalignment across generated multi-view images, leading to low-quality 3D generation results, such as incomplete geometric details and textural ghosting. Some methods are mainly optimized for the frontal perspective and exhibit poor robustness to oblique perspective inputs. In this paper, to tackle the above challenges, we propose a high-quality image-to-3D approach, named LSS3D, with learnable spatial shifting to explicitly and effectively handle the multiview inconsistencies and non-frontal input view. Specifically, we assign learnable spatial shifting parameters to each view, and adjust each view towards a spatially consistent target, guided by the reconstructed mesh, resulting in high-quality 3D generation with more complete geometric details and clean textures. Besides, we include the input view as an extra constraint for the optimization, further enhancing robustness to non-frontal input angles, especially for elevated viewpoint inputs. We also provide a comprehensive quantitative evaluation pipeline that can contribute to the community in performance comparisons. Extensive experiments demonstrate that our method consistently achieves leading results in both geometric and texture evaluation metrics across more flexible input viewpoints.
Authors: Jiaqi Wu, Yaosen Chen, Shuyuan Zhu
Abstract: Multi-view image generation holds significant application value in computer vision, particularly in domains like 3D reconstruction, virtual reality, and augmented reality. Most existing methods, which rely on extending single images, face notable computational challenges in maintaining cross-view consistency and generating high-resolution outputs. To address these issues, we propose the Geometry-guided Multi-View Diffusion Model, which incorporates mechanisms for extracting multi-view geometric information and adjusting the intensity of geometric features to generate images that are both consistent across views and rich in detail. Specifically, we design a multi-view geometry information extraction module that leverages depth maps, normal maps, and foreground segmentation masks to construct a shared geometric structure, ensuring shape and structural consistency across different views. To enhance consistency and detail restoration during generation, we develop a decoupled geometry-enhanced attention mechanism that strengthens feature focus on key geometric details, thereby improving overall image quality and detail preservation. Furthermore, we apply an adaptive learning strategy that fine-tunes the model to better capture spatial relationships and visual coherence between the generated views, ensuring realistic results. Our model also incorporates an iterative refinement process that progressively improves the output quality through multiple stages of image generation. Finally, a dynamic geometry information intensity adjustment mechanism is proposed to adaptively regulate the influence of geometric data, optimizing overall quality while ensuring the naturalness of generated images. More details can be found on the project page: https://github.com/SobeyMIL/GeoMVD.com.
Authors: Nishant Vasantkumar Hegde, Aditi Agarwal, Minal Moharir
Abstract: Road safety is a critical global concern, with manual enforcement of helmet laws and vehicle safety standards (e.g., rear-view mirror presence) being resource-intensive and inconsistent. This paper presents an AI-powered system to automate traffic violation detection, significantly enhancing enforcement efficiency and road safety. The system leverages YOLOv8 for robust object detection and EasyOCR for license plate recognition. Trained on a custom dataset of annotated images (augmented for diversity), it identifies helmet non-compliance, the absence of rear-view mirrors on motorcycles, an innovative contribution to automated checks, and extracts vehicle registration numbers. A Streamlit-based interface facilitates real-time monitoring and violation logging. Advanced image preprocessing enhances license plate recognition, particularly under challenging conditions. Based on evaluation results, the model achieves an overall precision of 0.9147, a recall of 0.886, and a mean Average Precision (mAP@50) of 0.843. The mAP@50 95 of 0.503 further indicates strong detection capability under stricter IoU thresholds. This work demonstrates a practical and effective solution for automated traffic rule enforcement, with considerations for real-world deployment discussed.
Authors: Haozhe Liu, Ding Liu, Mingchen Zhuge, Zijian Zhou, Tian Xie, Sen He, Yukang Yang, Shuming Liu, Yuren Cong, Jiadong Guo, Hongyu Xu, Ke Xu, Kam-Woh Ng, Juan C. P\'erez, Juan-Manuel~P\'erez-R\'ua, Tao Xiang, Wei Liu, Shikun Liu, J\"urgen Schmidhuber
Abstract: We introduce MoS (Mixture of States), a novel fusion paradigm for multimodal diffusion models that merges modalities using flexible, state-based interactions. The core of MoS is a learnable, token-wise router that creates denoising timestep- and input-dependent interactions between modalities' hidden states, precisely aligning token-level features with the diffusion trajectory. This router sparsely selects the top-$k$ hidden states and is trained with an $\epsilon$-greedy strategy, efficiently selecting contextual features with minimal learnable parameters and negligible computational overhead. We validate our design with text-to-image generation (MoS-Image) and editing (MoS-Editing), which achieve state-of-the-art results. With only 3B to 5B parameters, our models match or surpass counterparts up to $4\times$ larger. These findings establish MoS as a flexible and compute-efficient paradigm for scaling multimodal diffusion models.
Authors: Peng Zhang, Zhihui Lai, Wenting Chen, Xu Wu, Heng Kong
Abstract: Medical vision-language pre-training (VLP) offers significant potential for advancing medical image understanding by leveraging paired image-report data. However, existing methods are limited by Fa}lse Negatives (FaNe) induced by semantically similar texts and insufficient fine-grained cross-modal alignment. To address these limitations, we propose FaNe, a semantic-enhanced VLP framework. To mitigate false negatives, we introduce a semantic-aware positive pair mining strategy based on text-text similarity with adaptive normalization. Furthermore, we design a text-conditioned sparse attention pooling module to enable fine-grained image-text alignment through localized visual representations guided by textual cues. To strengthen intra-modal discrimination, we develop a hard-negative aware contrastive loss that adaptively reweights semantically similar negatives. Extensive experiments on five downstream medical imaging benchmarks demonstrate that FaNe achieves state-of-the-art performance across image classification, object detection, and semantic segmentation, validating the effectiveness of our framework.
Authors: Ameen Ali, Tamim Zoabi, Lior Wolf
Abstract: Vision-language models (VLMs) frequently produce hallucinations in the form of descriptions of objects, attributes, or relations that do not exist in the image due to over-reliance on language priors and imprecise cross-modal grounding. We introduce Spectral Representation Filtering (SRF), a lightweight, training-free method to suppress such hallucinations by analyzing and correcting the covariance structure of the model's representations. SRF identifies low-rank hallucination modes through eigendecomposition of the covariance of the differences between features collected for truthful and hallucinatory captions, revealing structured biases in the feature space. A soft spectral filter then attenuates these modes in the feed-forward projection weights of deeper vLLM layers, equalizing feature variance while preserving semantic fidelity. Unlike decoding or retraining-based approaches, SRF operates entirely post-hoc, incurs zero inference overhead, and requires no architectural modifications. Across three families of VLMs (LLaVA-1.5, MiniGPT-4, and mPLUG-Owl2), SRF consistently reduces hallucination rates on MSCOCO, POPE-VQA, and other visual tasks benchmarks, achieving state-of-the-art faithfulness without degrading caption quality.
Authors: Dongdong Zhao, Qiben Xu, Ranxin Fang, Baogang Song
Abstract: Deep hashing improves retrieval efficiency through compact binary codes, yet it introduces severe and often overlooked privacy risks. The ability to reconstruct original training data from hash codes could lead to serious threats such as biometric forgery and privacy breaches. However, model inversion attacks specifically targeting deep hashing models remain unexplored, leaving their security implications unexamined. This research gap stems from the inaccessibility of genuine training hash codes and the highly discrete Hamming space, which prevents existing methods from adapting to deep hashing. To address these challenges, we propose DHMI, the first diffusion-based model inversion framework designed for deep hashing. DHMI first clusters an auxiliary dataset to derive semantic hash centers as surrogate anchors. It then introduces a surrogate-guided denoising optimization method that leverages a novel attack metric (fusing classification consistency and hash proximity) to dynamically select candidate samples. A cluster of surrogate models guides the refinement of these candidates, ensuring the generation of high-fidelity and semantically consistent images. Experiments on multiple datasets demonstrate that DHMI successfully reconstructs high-resolution, high-quality images even under the most challenging black-box setting, where no training hash codes are available. Our method outperforms the existing state-of-the-art model inversion attacks in black-box scenarios, confirming both its practical efficacy and the critical privacy risks inherent in deep hashing systems.
Authors: Huy M. Le, Dat Tien Nguyen, Phuc Binh Nguyen, Gia-Bao Le-Tran, Phu Truong Thien, Cuong Dinh, Minh Nguyen, Nga Nguyen, Thuy T. N. Nguyen, Huy Gia Ngo, Tan Nhat Nguyen, Binh T. Nguyen, Monojit Choudhury
Abstract: The Video Browser Showdown (VBS) challenges systems to deliver accurate results under strict time constraints. To meet this demand, we present Fusionista2.0, a streamlined video retrieval system optimized for speed and usability. All core modules were re-engineered for efficiency: preprocessing now relies on ffmpeg for fast keyframe extraction, optical character recognition uses Vintern-1B-v3.5 for robust multilingual text recognition, and automatic speech recognition employs faster-whisper for real-time transcription. For question answering, lightweight vision-language models provide quick responses without the heavy cost of large models. Beyond these technical upgrades, Fusionista2.0 introduces a redesigned user interface with improved responsiveness, accessibility, and workflow efficiency, enabling even non-expert users to retrieve relevant content rapidly. Evaluations demonstrate that retrieval time was reduced by up to 75% while accuracy and user satisfaction both increased, confirming Fusionista2.0 as a competitive and user-friendly system for large-scale video search.
Authors: Tolga Demiroglu (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey), Mehmet Ozan Unal (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey), Metin Ertas (Istanbul University, Istanbul, Turkey), Isa Yildirim (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey)
Abstract: We propose a prompt-conditioned framework built on MedSigLIP that injects textual priors via Feature-wise Linear Modulation (FiLM) and multi-scale pooling. Text prompts condition patch-token features on clinical intent, enabling data-efficient learning and rapid adaptation. The architecture combines global, local, and texture-aware pooling through separate regression heads fused by a lightweight MLP, trained with pairwise ranking loss. Evaluated on the LDCTIQA2023 (a public LDCT quality assessment challenge) with 1,000 training images, we achieve PLCC = 0.9575, SROCC = 0.9561, and KROCC = 0.8301, surpassing the top-ranked published challenge submissions and demonstrating the effectiveness of our prompt-guided approach.
Authors: Puzhen Wu, Hexin Dong, Yi Lin, Yihao Ding, Yifan Peng
Abstract: Radiology report generation from chest X-rays is an important task in artificial intelligence with the potential to greatly reduce radiologists' workload and shorten patient wait times. Despite recent advances, existing approaches often lack sufficient disease-awareness in visual representations and adequate vision-language alignment to meet the specialized requirements of medical image analysis. As a result, these models usually overlook critical pathological features on chest X-rays and struggle to generate clinically accurate reports. To address these limitations, we propose a novel dual-stage disease-aware framework for chest X-ray report generation. In Stage~1, our model learns Disease-Aware Semantic Tokens (DASTs) corresponding to specific pathology categories through cross-attention mechanisms and multi-label classification, while simultaneously aligning vision and language representations via contrastive learning. In Stage~2, we introduce a Disease-Visual Attention Fusion (DVAF) module to integrate disease-aware representations with visual features, along with a Dual-Modal Similarity Retrieval (DMSR) mechanism that combines visual and disease-specific similarities to retrieve relevant exemplars, providing contextual guidance during report generation. Extensive experiments on benchmark datasets (i.e., CheXpert Plus, IU X-ray, and MIMIC-CXR) demonstrate that our disease-aware framework achieves state-of-the-art performance in chest X-ray report generation, with significant improvements in clinical accuracy and linguistic quality.
Authors: Jingyao Li, Jingyun Wang, Molin Tan, Haochen Wang, Cilin Yan, Likun Shi, Jiayin Cai, Xiaolong Jiang, Yao Hu
Abstract: Cross-Video Reasoning (CVR) presents a significant challenge in video understanding, which requires simultaneous understanding of multiple videos to aggregate and compare information across groups of videos. Most existing video understanding benchmarks focus on single-video analysis, failing to assess the ability of multimodal large language models (MLLMs) to simultaneously reason over various videos. Recent benchmarks evaluate MLLMs' capabilities on multi-view videos that capture different perspectives of the same scene. However, their limited tasks hinder a thorough assessment of MLLMs in diverse real-world CVR scenarios. To this end, we introduce CrossVid, the first benchmark designed to comprehensively evaluate MLLMs' spatial-temporal reasoning ability in cross-video contexts. Firstly, CrossVid encompasses a wide spectrum of hierarchical tasks, comprising four high-level dimensions and ten specific tasks, thereby closely reflecting the complex and varied nature of real-world video understanding. Secondly, CrossVid provides 5,331 videos, along with 9,015 challenging question-answering pairs, spanning single-choice, multiple-choice, and open-ended question formats. Through extensive experiments on various open-source and closed-source MLLMs, we observe that Gemini-2.5-Pro performs best on CrossVid, achieving an average accuracy of 50.4%. Notably, our in-depth case study demonstrates that most current MLLMs struggle with CVR tasks, primarily due to their inability to integrate or compare evidence distributed across multiple videos for reasoning. These insights highlight the potential of CrossVid to guide future advancements in enhancing MLLMs' CVR capabilities.
Authors: Ruixun Liu, Bowen Fu, Jiayi Song, Kaiyu Li, Wanchen Li, Lanxuan Xue, Hui Qiao, Weizhan Zhang, Deyu Meng, Xiangyong Cao
Abstract: Ultra-high-resolution (UHR) remote sensing (RS) images offer rich fine-grained information but also present challenges in effective processing. Existing dynamic resolution and token pruning methods are constrained by a passive perception paradigm, suffering from increased redundancy when obtaining finer visual inputs. In this work, we explore a new active perception paradigm that enables models to revisit information-rich regions. First, we present LRS-GRO, a large-scale benchmark dataset tailored for active perception in UHR RS processing, encompassing 17 question types across global, region, and object levels, annotated via a semi-automatic pipeline. Building on LRS-GRO, we propose ZoomEarth, an adaptive cropping-zooming framework with a novel Region-Guided reward that provides fine-grained guidance. Trained via supervised fine-tuning (SFT) and Group Relative Policy Optimization (GRPO), ZoomEarth achieves state-of-the-art performance on LRS-GRO and, in the zero-shot setting, on three public UHR remote sensing benchmarks. Furthermore, ZoomEarth can be seamlessly integrated with downstream models for tasks such as cloud removal, denoising, segmentation, and image editing through simple tool interfaces, demonstrating strong versatility and extensibility.
Authors: Yaxuan Jiao, Qing Xu, Yuxiang Luo, Xiangjian He, Zhen Chen, Wenting Duan
Abstract: Medical image segmentation is essential for clinical diagnosis and treatment planning. Although transformer-based methods have achieved remarkable results, their high computational cost hinders clinical deployment. To address this issue, we propose TM-UNet, a novel lightweight framework that integrates token sequence modeling with an efficient memory mechanism for efficient medical segmentation. Specifically, we introduce a multi-scale token-memory (MSTM) block that transforms 2D spatial features into token sequences through strategic spatial scanning, leveraging matrix memory cells to selectively retain and propagate discriminative contextual information across tokens. This novel token-memory mechanism acts as a dynamic knowledge store that captures long-range dependencies with linear complexity, enabling efficient global reasoning without redundant computation. Our MSTM block further incorporates exponential gating to identify token effectiveness and multi-scale contextual extraction via parallel pooling operations, enabling hierarchical representation learning without computational overhead. Extensive experiments demonstrate that TM-UNet outperforms state-of-the-art methods across diverse medical segmentation tasks with substantially reduced computation cost. The code is available at https://github.com/xq141839/TM-UNet.
Authors: Shuochen Chang, Xiaofeng Zhang, Qingyang Liu, Li Niu
Abstract: Diffusion-based multimodal large language models (Diffusion MLLMs) have recently demonstrated impressive non-autoregressive generative capabilities across vision-and-language tasks. However, Diffusion MLLMs exhibit substantially slower inference than autoregressive models: Each denoising step employs full bidirectional self-attention over the entire sequence, resulting in cubic decoding complexity that becomes computationally impractical with thousands of visual tokens. To address this challenge, we propose D$^{3}$ToM, a Decider-guided dynamic token merging method that dynamically merges redundant visual tokens at different denoising steps to accelerate inference in Diffusion MLLMs. At each denoising step, D$^{3}$ToM uses decider tokens-the tokens generated in the previous denoising step-to build an importance map over all visual tokens. Then it maintains a proportion of the most salient tokens and merges the remainder through similarity-based aggregation. This plug-and-play module integrates into a single transformer layer, physically shortening the visual token sequence for all subsequent layers without altering model parameters. Moreover, D$^{3}$ToM employs a merge ratio that dynamically varies with each denoising step, aligns with the native decoding process of Diffusion MLLMs, achieving superior performance under equivalent computational budgets. Extensive experiments show that D$^{3}$ToM accelerates inference while preserving competitive performance. The code is released at https://github.com/bcmi/D3ToM-Diffusion-MLLM.
Authors: Andrea Bertogalli, Giacomo Boracchi, Luca Magri
Abstract: We present a novel multi-modal extrinsic calibration framework designed to simultaneously estimate the relative poses between event cameras, LiDARs, and RGB cameras, with particular focus on the challenging event camera calibration. Core of our approach is a novel 3D calibration target, specifically designed and constructed to be concurrently perceived by all three sensing modalities. The target encodes features in planes, ChArUco, and active LED patterns, each tailored to the unique characteristics of LiDARs, RGB cameras, and event cameras respectively. This unique design enables a one-shot, joint extrinsic calibration process, in contrast to existing approaches that typically rely on separate, pairwise calibrations. Our calibration pipeline is designed to accurately calibrate complex vision systems in the context of autonomous driving, where precise multi-sensor alignment is critical. We validate our approach through an extensive experimental evaluation on a custom built dataset, recorded with an advanced autonomous driving sensor setup, confirming the accuracy and robustness of our method.
Authors: Chi Liu, Jincheng Liu, Congcong Zhu, Minghao Wang, Sheng Shen, Jia Gu, Tianqing Zhu, Wanlei Zhou
Abstract: Developing Medical AI relies on large datasets and easily suffers from data scarcity. Generative data augmentation (GDA) using AI generative models offers a solution to synthesize realistic medical images. However, the bias in GDA is often underestimated in medical domains, with concerns about the risk of introducing detrimental features generated by AI and harming downstream tasks. This paper identifies the frequency misalignment between real and synthesized images as one of the key factors underlying unreliable GDA and proposes the Frequency Recalibration (FreRec) method to reduce the frequency distributional discrepancy and thus improve GDA. FreRec involves (1) Statistical High-frequency Replacement (SHR) to roughly align high-frequency components and (2) Reconstructive High-frequency Mapping (RHM) to enhance image quality and reconstruct high-frequency details. Extensive experiments were conducted in various medical datasets, including brain MRIs, chest X-rays, and fundus images. The results show that FreRec significantly improves downstream medical image classification performance compared to uncalibrated AI-synthesized samples. FreRec is a standalone post-processing step that is compatible with any generative model and can integrate seamlessly with common medical GDA pipelines.
Authors: Qifeng Chen, Jiarun Liu, Rengan Xie, Tao Tang, Sicong Du, Yiru Zhao, Yuchi Huo, Sheng Yang
Abstract: Recent GS-based rendering has made significant progress for LiDAR, surpassing Neural Radiance Fields (NeRF) in both quality and speed. However, these methods exhibit artifacts in extrapolated novel view synthesis due to the incomplete reconstruction from single traversal scans. To address this limitation, we present LiDAR-GS++, a LiDAR Gaussian Splatting reconstruction method enhanced by diffusion priors for real-time and high-fidelity re-simulation on public urban roads. Specifically, we introduce a controllable LiDAR generation model conditioned on coarsely extrapolated rendering to produce extra geometry-consistent scans and employ an effective distillation mechanism for expansive reconstruction. By extending reconstruction to under-fitted regions, our approach ensures global geometric consistency for extrapolative novel views while preserving detailed scene surfaces captured by sensors. Experiments on multiple public datasets demonstrate that LiDAR-GS++ achieves state-of-the-art performance for both interpolated and extrapolated viewpoints, surpassing existing GS and NeRF-based methods.
Authors: Xi Ding, Lei Wang, Piotr Koniusz, Yongsheng Gao
Abstract: Real-world visual data rarely presents as isolated, static instances. Instead, it often evolves gradually over time through variations in pose, lighting, object state, or scene context. However, conventional classifiers are typically trained under the assumption of temporal independence, limiting their ability to capture such dynamics. We propose a simple yet effective framework that equips standard feedforward classifiers with temporal reasoning, all without modifying model architectures or introducing recurrent modules. At the heart of our approach is a novel Support-Exemplar-Query (SEQ) learning paradigm, which structures training data into temporally coherent trajectories. These trajectories enable the model to learn class-specific temporal prototypes and align prediction sequences via a differentiable soft-DTW loss. A multi-term objective further promotes semantic consistency and temporal smoothness. By interpreting input sequences as evolving feature trajectories, our method introduces a strong temporal inductive bias through loss design alone. This proves highly effective in both static and temporal tasks: it enhances performance on fine-grained and ultra-fine-grained image classification, and delivers precise, temporally consistent predictions in video anomaly detection. Despite its simplicity, our approach bridges static and temporal learning in a modular and data-efficient manner, requiring only a simple classifier on top of pre-extracted features.
Authors: Sepehr Kazemi Ranjbar, Kumail Alhamoud, Marzyeh Ghassemi
Abstract: Vision-Language Models (VLMs) struggle with negation. Given a prompt like "retrieve (or generate) a street scene without pedestrians," they often fail to respect the "not." Existing methods address this limitation by fine-tuning on large negation datasets, but such retraining often compromises the model's zero-shot performance on affirmative prompts. We show that the embedding space of VLMs, such as CLIP, can be divided into semantically consistent subspaces. Based on this property, we propose a training-free framework that models negation as a subspace in the joint embedding space rather than a single point (Figure 1). To find the matching image for a caption such as "A but not N," we construct two spherical caps around the embeddings of A and N, and we score images by the central direction of the region that is close to A and far from N. Across retrieval, MCQ, and text-to-image tasks, our method improves negation understanding by about 30% on average over prior methods. It closes the gap between affirmative and negated prompts while preserving the zero-shot performance that fine-tuned models fail to maintain. Code will be released upon publication.
Authors: Sajjad Pakdamansavoji, Kumar Vaibhav Jha, Baher Abdulhai, James H Elder
Abstract: Accurate turning movement counts at intersections are important for signal control, traffic management and urban planning. Computer vision systems for automatic turning movement counts typically rely on visual analysis in the image plane of an infrastructure camera. Here we explore potential advantages of back-projecting vehicles detected in one or more infrastructure cameras to the ground plane for analysis in real-world 3D coordinates. For single-camera systems we find that back-projection yields more accurate trajectory classification and turning movement counts. We further show that even higher accuracy can be achieved through weak fusion of back-projected detections from multiple cameras. These results suggeest that traffic should be analyzed on the ground plane, not the image plane
Authors: Asmit Bandyopadhyay, Anindita Das Bhattacharjee, Rakesh Das
Abstract: Hyperspectral image (HSI) classification faces critical challenges, including high spectral dimensionality, complex spectral-spatial correlations, and limited training samples with severe class imbalance. While CNNs excel at local feature extraction and transformers capture long-range dependencies, their isolated application yields suboptimal results due to quadratic complexity and insufficient inductive biases. We propose CLAReSNet (Convolutional Latent Attention Residual Spectral Network), a hybrid architecture that integrates multi-scale convolutional extraction with transformer-style attention via an adaptive latent bottleneck. The model employs a multi-scale convolutional stem with deep residual blocks and an enhanced Convolutional Block Attention Module for hierarchical spatial features, followed by spectral encoder layers combining bidirectional RNNs (LSTM/GRU) with Multi-Scale Spectral Latent Attention (MSLA). MSLA reduces complexity from $\mathcal{O}(T^2D)$ to $\mathcal{O}(T\log(T)D)$ by adaptive latent token allocation (8-64 tokens) that scales logarithmically with the sequence length. Hierarchical cross-attention fusion dynamically aggregates multi-level representations for robust classification. Experiments conducted on the Indian Pines and Salinas datasets show state-of-the-art performance, achieving overall accuracies of 99.71% and 99.96%, significantly surpassing HybridSN, SSRN, and SpectralFormer. The learned embeddings exhibit superior inter-class separability and compact intra-class clustering, validating CLAReSNet's effectiveness under limited samples and severe class imbalance.
Authors: Michael Yang, Shijian Deng, William T. Doan, Kai Wang, Tianyu Yang, Harsh Singh, Yapeng Tian
Abstract: Conventional, classification-based AI-generated image detection methods cannot explain why an image is considered real or AI-generated in a way a human expert would, which reduces the trustworthiness and persuasiveness of these detection tools for real-world applications. Leveraging Multimodal Large Language Models (MLLMs) has recently become a trending solution to this issue. Further, to evaluate the quality of generated explanations, a common approach is to adopt an "MLLM as a judge" methodology to evaluate explanations generated by other MLLMs. However, how well those MLLMs perform when judging explanations for AI-generated image detection generated by themselves or other MLLMs has not been well studied. We therefore propose \textbf{XAIGID-RewardBench}, the first benchmark designed to evaluate the ability of current MLLMs to judge the quality of explanations about whether an image is real or AI-generated. The benchmark consists of approximately 3,000 annotated triplets sourced from various image generation models and MLLMs as policy models (detectors) to assess the capabilities of current MLLMs as reward models (judges). Our results show that the current best reward model scored 88.76\% on this benchmark (while human inter-annotator agreement reaches 98.30\%), demonstrating that a visible gap remains between the reasoning abilities of today's MLLMs and human-level performance. In addition, we provide an analysis of common pitfalls that these models frequently encounter. Code and benchmark are available at https://github.com/RewardBench/XAIGID-RewardBench.
Authors: Yiqing Shen, Mathias Unberath
Abstract: Visual reasoning may require models to interpret images and videos and respond to implicit text queries across diverse output formats, from pixel-level segmentation masks to natural language descriptions. Existing approaches rely on supervised fine-tuning with task-specific architectures. For example, reasoning segmentation, grounding, summarization, and visual question answering each demand distinct model designs and training, preventing unified solutions and limiting cross-task and cross-modality generalization. Hence, we propose DT-R1, a reinforcement learning framework that trains large language models to construct digital twin representations of complex multi-modal visual inputs and then reason over these high-level representations as a unified approach to visual reasoning. Specifically, we train DT-R1 using GRPO with a novel reward that validates both structural integrity and output accuracy. Evaluations in six visual reasoning benchmarks, covering two modalities and four task types, demonstrate that DT-R1 consistently achieves improvements over state-of-the-art task-specific models. DT-R1 opens a new direction where visual reasoning emerges from reinforcement learning with digital twin representations.
Authors: Yiqing Shen, Mathias Unberath
Abstract: Reasoning segmentation enables open-set object segmentation via implicit text queries, therefore serving as a foundation for embodied agents that should operate autonomously in real-world environments. However, existing methods for reasoning segmentation require multimodal large language models with billions of parameters that exceed the computational capabilities of edge devices that typically deploy the embodied AI systems. Distillation offers a pathway to compress these models while preserving their capabilities. Yet, existing distillation approaches fail to transfer the multi-step reasoning capabilities that reasoning segmentation demands, as they focus on matching output predictions and intermediate features rather than preserving reasoning chains. The emerging paradigm of reasoning over digital twin representations presents an opportunity for more effective distillation by re-framing the problem. Consequently, we propose FastReasonSeg, which employs digital twin representations that decouple perception from reasoning to enable more effective distillation. Our distillation scheme first relies on supervised fine-tuning on teacher-generated reasoning chains. Then it is followed by reinforcement fine-tuning with joint rewards evaluating both segmentation accuracy and reasoning quality alignment. Experiments on two video (JiTBench, RVTBench) and two image benchmarks (ReasonSeg, LLM-Seg40K) demonstrate that our FastReasonSeg achieves state-of-the-art reasoning segmentation performance. Moreover, the distilled 0.6B variant outperforms models with 20 times more parameters while achieving 7.79 FPS throughput with only 2.1GB memory consumption. This efficiency enables deployment in resource-constrained environments to enable real-time reasoning segmentation.
Authors: Chamuditha Jayanga Galappaththige, Jason Lai, Lloyd Windrim, Donald Dansereau, Niko S\"underhauf, Dimity Miller
Abstract: Online Scene Change Detection (SCD) is an extremely challenging problem that requires an agent to detect relevant changes on the fly while observing the scene from unconstrained viewpoints. Existing online SCD methods are significantly less accurate than offline approaches. We present the first online SCD approach that is pose-agnostic, label-free, and ensures multi-view consistency, while operating at over 10 FPS and achieving new state-of-the-art performance, surpassing even the best offline approaches. Our method introduces a new self-supervised fusion loss to infer scene changes from multiple cues and observations, PnP-based fast pose estimation against the reference scene, and a fast change-guided update strategy for the 3D Gaussian Splatting scene representation. Extensive experiments on complex real-world datasets demonstrate that our approach outperforms both online and offline baselines.
Authors: Yiqing Shen, Chenxiao Fan, Chenjia Li, Mathias Unberath
Abstract: The goal of text-to-video retrieval is to search large databases for relevant videos based on text queries. Existing methods have progressed to handling explicit queries where the visual content of interest is described explicitly; however, they fail with implicit queries where identifying videos relevant to the query requires reasoning. We introduce reasoning text-to-video retrieval, a paradigm that extends traditional retrieval to process implicit queries through reasoning while providing object-level grounding masks that identify which entities satisfy the query conditions. Instead of relying on vision-language models directly, we propose representing video content as digital twins, i.e., structured scene representations that decompose salient objects through specialist vision models. This approach is beneficial because it enables large language models to reason directly over long-horizon video content without visual token compression. Specifically, our two-stage framework first performs compositional alignment between decomposed sub-queries and digital twin representations for candidate identification, then applies large language model-based reasoning with just-in-time refinement that invokes additional specialist models to address information gaps. We construct a benchmark of 447 manually created implicit queries with 135 videos (ReasonT2VBench-135) and another more challenging version of 1000 videos (ReasonT2VBench-1000). Our method achieves 81.2% R@1 on ReasonT2VBench-135, outperforming the strongest baseline by greater than 50 percentage points, and maintains 81.7% R@1 on the extended configuration while establishing state-of-the-art results in three conventional benchmarks (MSR-VTT, MSVD, and VATEX).
Authors: Ansh Makwe, Akansh Agrawal, Prateek Jain, Akshan Agrawal, Priyanka Bagade
Abstract: Medical image analysis for complex tasks such as severity grading and disease subtype classification poses significant challenges due to intricate and similar visual patterns among classes, scarcity of labeled data, and variability in expert interpretations. Despite the usefulness of existing attention-based models in capturing complex visual patterns for medical image classification, underlying architectures often face challenges in effectively distinguishing subtle classes since they struggle to capture inter-class similarity and intra-class variability, resulting in incorrect diagnosis. To address this, we propose AGGRNet framework to extract informative and non-informative features to effectively understand fine-grained visual patterns and improve classification for complex medical image analysis tasks. Experimental results show that our model achieves state-of-the-art performance on various medical imaging datasets, with the best improvement up to 5% over SOTA models on the Kvasir dataset.
Authors: Shabnam Sodagari, Tommy Long
Abstract: The objective of this study is to diagnose and differentiate kidney stones, cysts, and tumors using Computed Tomography (CT) images of the kidney. This study leverages a hybrid quantum-classical framework in this regard. We combine a pretrained ResNet50 encoder, with a Quantum Convolutional Neural Network (QCNN) to explore quantum-assisted diagnosis. We pre-process the kidney images using denoising and contrast limited adaptive histogram equalization to enhance feature extraction. We address class imbalance through data augmentation and weighted sampling. Latent features extracted by the encoder are transformed into qubits via angle encoding and processed by a QCNN. The model is evaluated on both 8-qubit and 12-qubit configurations. Both architectures achieved rapid convergence with stable learning curves and high consistency between training and validation performance. The models reached a test accuracy of 0.99, with the 12-qubit configuration providing improvements in overall recall and precision, particularly for Cyst and Tumor detection, where it achieved perfect recall for Cysts and a tumor F1-score of 0.9956. Confusion matrix analysis further confirmed reliable classification behavior across all classes, with very few misclassifications. Results demonstrate that integrating classical pre-processing and deep feature extraction with quantum circuits enhances medical diagnostic performance.
Authors: Divake Kumar, Patrick Poggi, Sina Tayebati, Devashri Naik, Nilesh Ahuja, Amit Ranjan Trivedi
Abstract: Most estimators collapse all uncertainty modes into a single confidence score, preventing reliable reasoning about when to allocate more compute or adjust inference. We introduce Uncertainty-Guided Inference-Time Selection, a lightweight inference time framework that disentangles aleatoric (data-driven) and epistemic (model-driven) uncertainty directly in deep feature space. Aleatoric uncertainty is estimated using a regularized global density model, while epistemic uncertainty is formed from three complementary components that capture local support deficiency, manifold spectral collapse, and cross-layer feature inconsistency. These components are empirically orthogonal and require no sampling, no ensembling, and no additional forward passes. We integrate the decomposed uncertainty into a distribution free conformal calibration procedure that yields significantly tighter prediction intervals at matched coverage. Using these components for uncertainty guided adaptive model selection reduces compute by approximately 60 percent on MOT17 with negligible accuracy loss, enabling practical self regulating visual inference. Additionally, our ablation results show that the proposed orthogonal uncertainty decomposition consistently yields higher computational savings across all MOT17 sequences, improving margins by 13.6 percentage points over the total-uncertainty baseline.
Authors: Xu Yang, Gady Agam
Abstract: We introduce MSLoRA, a backbone-agnostic, parameter-efficient adapter that reweights feature responses rather than re-tuning the underlying backbone. Existing low-rank adaptation methods are mostly confined to vision transformers (ViTs) and struggle to generalize across architectures. MSLoRA unifies adaptation for both convolutional neural networks (CNNs) and ViTs by combining a low-rank linear projection with a multi-scale nonlinear transformation that jointly modulates spatial and channel attention. The two components are fused through pointwise multiplication and a residual connection, yielding a lightweight module that shifts feature attention while keeping pretrained weights frozen. Extensive experiments demonstrate that MSLoRA consistently improves transfer performance on classification, detection, and segmentation tasks with roughly less than 5\% of backbone parameters. The design further enables stable optimization, fast convergence, and strong cross-architecture generalization. By reweighting rather than re-tuning, MSLoRA provides a simple and universal approach for efficient adaptation of frozen vision backbones.
Authors: Hyunki Seong, Seongwoo Moon, Hojin Ahn, Jehun Kang, David Hyunchul Shim
Abstract: Exploring open-world situations in an end-to-end manner is a promising yet challenging task due to the need for strong generalization capabilities. In particular, end-to-end autonomous driving in unstructured outdoor environments often encounters conditions that were unfamiliar during training. In this work, we present Vision-Language Action Retrieval (VLA-R), an open-world end-to-end autonomous driving (OW-E2EAD) framework that integrates open-world perception with a novel vision-action retrieval paradigm. We leverage a frozen vision-language model for open-world detection and segmentation to obtain multi-scale, prompt-guided, and interpretable perception features without domain-specific tuning. A Q-Former bottleneck aggregates fine-grained visual representations with language-aligned visual features, bridging perception and action domains. To learn transferable driving behaviors, we introduce a vision-action contrastive learning scheme that aligns vision-language and action embeddings for effective open-world reasoning and action retrieval. Our experiments on a real-world robotic platform demonstrate strong generalization and exploratory performance in unstructured, unseen environments, even with limited data. Demo videos are provided in the supplementary material.
Authors: Xi Xiao, Zhuxuanzi Wang, Mingqiao Mo, Chen Liu, Chenrui Ma, Yanshu Li, Smita Krishnaswamy, Xiao Wang, Tianyang Wang
Abstract: The deployment of automated pavement defect detection is often hindered by poor cross-domain generalization. Supervised detectors achieve strong in-domain accuracy but require costly re-annotation for new environments, while standard self-supervised methods capture generic features and remain vulnerable to domain shift. We propose \ours, a self-supervised framework that \emph{visually probes} target domains without labels. \ours introduces a Self-supervised Prompt Enhancement Module (SPEM), which derives defect-aware prompts from unlabeled target data to guide a frozen ViT backbone, and a Domain-Aware Prompt Alignment (DAPA) objective, which aligns prompt-conditioned source and target representations. Experiments on four challenging benchmarks show that \ours consistently outperforms strong supervised, self-supervised, and adaptation baselines, achieving robust zero-shot transfer, improved resilience to domain variations, and high data efficiency in few-shot adaptation. These results highlight self-supervised prompting as a practical direction for building scalable and adaptive visual inspection systems. Source code is publicly available: https://github.com/xixiaouab/PROBE/tree/main
Authors: Xinrui Li, Qi Cai, Yuanxin Wu
Abstract: Structure from Motion (SfM) is a critical task in computer vision, aiming to recover the 3D scene structure and camera motion from a sequence of 2D images. The recent pose-only imaging geometry decouples 3D coordinates from camera poses and demonstrates significantly better SfM performance through pose adjustment. Continuing the pose-only perspective, this paper explores the critical relationship between the scene structures, rotation and translation. Notably, the translation can be expressed in terms of rotation, allowing us to condense the imaging geometry representation onto the rotation manifold. A rotation-only optimization framework based on reprojection error is proposed for both two-view and multi-view scenarios. The experiment results demonstrate superior accuracy and robustness performance over the current state-of-the-art rotation estimation methods, even comparable to multiple bundle adjustment iteration results. Hopefully, this work contributes to even more accurate, efficient and reliable 3D visual computing.
Authors: Wenjie Li, Jinglei Shi, Jin Han, Heng Guo, Zhanyu Ma
Abstract: Clean images are crucial for visual tasks such as small object detection, especially at high resolutions. However, real-world images are often degraded by adverse weather, and weather restoration methods may sacrifice high-frequency details critical for analyzing small objects. A natural solution is to apply super-resolution (SR) after weather removal to recover both clarity and fine structures. However, simply cascading restoration and SR struggle to bridge their inherent conflict: removal aims to remove high-frequency weather-induced noise, while SR aims to hallucinate high-frequency textures from existing details, leading to inconsistent restoration contents. In this paper, we take deraining as a case study and propose DHGM, a Diffusion-based High-frequency Guided Model for generating clean and high-resolution images. DHGM integrates pre-trained diffusion priors with high-pass filters to simultaneously remove rain artifacts and enhance structural details. Extensive experiments demonstrate that DHGM achieves superior performance over existing methods, with lower costs.
Authors: Nuolin Sun, Linyuan Wang, Haonan Wei, Lei Li, Bin Yan
Abstract: ResNet has achieved tremendous success in computer vision through its residual connection mechanism. ResNet can be viewed as a discretized form of ordinary differential equations (ODEs). From this perspective, the multiple residual blocks within a single ResNet stage essentially perform multi-step discrete iterations of the feature transformation for that stage. The recently proposed flow matching model, MeanFlow, enables one-step generative modeling by learning the mean velocity field to transform distributions. Inspired by this, we propose MeanFlow-Incubated ResNet (MFI-ResNet), which employs a compression-expansion strategy to jointly improve parameter efficiency and discriminative performance. In the compression phase, we simplify the multi-layer structure within each ResNet stage to one or two MeanFlow modules to construct a lightweight meta model. In the expansion phase, we apply a selective incubation strategy to the first three stages, expanding them to match the residual block configuration of the baseline ResNet model, while keeping the last stage in MeanFlow form, and fine-tune the incubated model. Experimental results show that on CIFAR-10 and CIFAR-100 datasets, MFI-ResNet achieves remarkable parameter efficiency, reducing parameters by 46.28% and 45.59% compared to ResNet-50, while still improving accuracy by 0.23% and 0.17%, respectively. This demonstrates that generative flow-fields can effectively characterize the feature transformation process in ResNet, providing a new perspective for understanding the relationship between generative modeling and discriminative learning.
Authors: Jingqi Xu, Jingxi Lu, Chenghao Li, Sreetama Sarkar, Souvik Kundu, Peter A. Beerel
Abstract: Vision-Language Models (VLMs) have achieved remarkable progress in multimodal reasoning and generation, yet their high computational demands remain a major challenge. Diffusion Vision-Language Models (DVLMs) are particularly attractive because they enable parallel token decoding, but the large number of visual tokens still significantly hinders their inference efficiency. While visual token pruning has been extensively studied for autoregressive VLMs (AVLMs), it remains largely unexplored for DVLMs. In this work, we propose RedVTP, a response-driven visual token pruning strategy that leverages the inference dynamics of DVLMs. Our method estimates visual token importance using attention from the masked response tokens. Based on the observation that these importance scores remain consistent across steps, RedVTP prunes the less important visual tokens from the masked tokens after the first inference step, thereby maximizing inference efficiency. Experiments show that RedVTP improves token generation throughput of LLaDA-V and LaViDa by up to 186% and 28.05%, respectively, and reduces inference latency by up to 64.97% and 21.87%, without compromising-and in some cases improving-accuracy.
Authors: Xilai Li, Xiaosong Li, Weijun Jiang
Abstract: Multi-modality image fusion enhances scene perception by combining complementary information. Unified models aim to share parameters across modalities for multi-modality image fusion, but large modality differences often cause gradient conflicts, limiting performance. Some methods introduce modality-specific encoders to enhance feature perception and improve fusion quality. However, this strategy reduces generalisation across different fusion tasks. To overcome this limitation, we propose a unified multi-modality image fusion framework based on channel perturbation and pre-trained knowledge integration (UP-Fusion). To suppress redundant modal information and emphasize key features, we propose the Semantic-Aware Channel Pruning Module (SCPM), which leverages the semantic perception capability of a pre-trained model to filter and enhance multi-modality feature channels. Furthermore, we proposed the Geometric Affine Modulation Module (GAM), which uses original modal features to apply affine transformations on initial fusion features to maintain the feature encoder modal discriminability. Finally, we apply a Text-Guided Channel Perturbation Module (TCPM) during decoding to reshape the channel distribution, reducing the dependence on modality-specific channels. Extensive experiments demonstrate that the proposed algorithm outperforms existing methods on both multi-modality image fusion and downstream tasks.
Authors: ANK Zaman, Prosenjit Chatterjee, Rajat Sharma
Abstract: A long road trip is fun for drivers. However, a long drive for days can be tedious for a driver to accommodate stringent deadlines to reach distant destinations. Such a scenario forces drivers to drive extra miles, utilizing extra hours daily without sufficient rest and breaks. Once a driver undergoes such a scenario, it occasionally triggers drowsiness during driving. Drowsiness in driving can be life-threatening to any individual and can affect other drivers' safety; therefore, a real-time detection system is needed. To identify fatigued facial characteristics in drivers and trigger the alarm immediately, this research develops a real-time driver drowsiness detection system utilizing deep convolutional neural networks (DCNNs) and OpenCV.Our proposed and implemented model takes real- time facial images of a driver using a live camera and utilizes a Python-based library named OpenCV to examine the facial images for facial landmarks like sufficient eye openings and yawn-like mouth movements. The DCNNs framework then gathers the data and utilizes a per-trained model to detect the drowsiness of a driver using facial landmarks. If the driver is identified as drowsy, the system issues a continuous alert in real time, embedded in the Smart Car technology.By potentially saving innocent lives on the roadways, the proposed technique offers a non-invasive, inexpensive, and cost-effective way to identify drowsiness. Our proposed and implemented DCNNs embedded drowsiness detection model successfully react with NTHU-DDD dataset and Yawn-Eye-Dataset with drowsiness detection classification accuracy of 99.6% and 97% respectively.
Authors: Jiahe Qian, Yuhao Shen, Zhangtianyi Chen, Juexiao Zhou, Peisong Wang
Abstract: Medical visual question answering could support clinical decision making, yet current systems often fail under domain shift and produce answers that are weakly grounded in image evidence. This reliability gap arises when models attend to spurious regions and when retraining or additional labels are impractical at deployment time. We address this setting with CoTBox-TTT, an evidence-first test-time training approach that adapts a vision-language model at inference while keeping all backbones frozen. The method updates only a small set of continuous soft prompts. It identifies question-relevant regions through a visual chain-of-thought signal and encourages answer consistency across the original image and a localized crop. The procedure is label free, and plug and play with diverse backbones. Experiments on medical VQA show that the approach is practical for real deployments. For instance, adding CoTBox-TTT to LLaVA increases closed-ended accuracy by 12.3% on pathVQA.
Authors: Zhanheng Nie, Chenghan Fu, Daoze Zhang, Junxian Wu, Wanxian Guan, Pengjie Wang, Jian Xu, Bo Zheng
Abstract: The rapid growth of e-commerce calls for multimodal models that comprehend rich visual and textual product information. Although recent multimodal large language models (MLLMs) for product understanding exhibit strong capability in representation learning for e-commerce, they still face three challenges: (i) the modality imbalance induced by modality mixed training; (ii) underutilization of the intrinsic alignment relationships among visual and textual information within a product; and (iii) limited handling of noise in e-commerce multimodal data. To address these, we propose MOON2.0, a dynamic modality-balanced multimodal representation learning framework for e-commerce product understanding. MOON2.0 comprises: (1) a Modality-driven Mixture-of-Experts (MoE) module that adaptively processes input samples by their modality composition, enabling Multimodal Joint Learning to mitigate the modality imbalance; (2) a Dual-level Alignment method to better leverage semantic alignment properties inside individual products; and (3) an MLLM-based Image-text Co-augmentation strategy that integrates textual enrichment with visual expansion, coupled with Dynamic Sample Filtering to improve training data quality. We further introduce MBE2.0, a co-augmented multimodal representation benchmark for e-commerce representation learning and evaluation. Experiments show that MOON2.0 delivers state-of-the-art zero-shot performance on MBE2.0 and multiple public datasets. Furthermore, attention-based heatmap visualization provides qualitative evidence of improved multimodal alignment of MOON2.0.
Authors: Xiaoyu Lin, Aniket Ghorpade, Hansheng Zhu, Justin Qiu, Dea Rrozhani, Monica Lama, Mick Yang, Zixuan Bian, Ruohan Ren, Alan B. Hong, Jiatao Gu, Chris Callison-Burch
Abstract: With the rapid adoption of multimodal large language models (MLLMs) across diverse applications, there is a pressing need for task-centered, high-quality training data. A key limitation of current training datasets is their reliance on sparse annotations mined from the Internet or entered via manual typing that capture only a fraction of an image's visual content. Dense annotations are more valuable but remain scarce. Traditional text-based annotation pipelines are poorly suited for creating dense annotations: typing limits expressiveness, slows annotation speed, and underrepresents nuanced visual features, especially in specialized areas such as multicultural imagery and 3D asset annotation. In this paper, we present DenseAnnotate, an audio-driven online annotation platform that enables efficient creation of dense, fine-grained annotations for images and 3D assets. Annotators narrate observations aloud while synchronously linking spoken phrases to image regions or 3D scene parts. Our platform incorporates speech-to-text transcription and region-of-attention marking. To demonstrate the effectiveness of DenseAnnotate, we conducted case studies involving over 1,000 annotators across two domains: culturally diverse images and 3D scenes. We curate a human-annotated multi-modal dataset of 3,531 images, 898 3D scenes, and 7,460 3D objects, with audio-aligned dense annotations in 20 languages, including 8,746 image captions, 2,000 scene captions, and 19,000 object captions. Models trained on this dataset exhibit improvements of 5% in multilingual, 47% in cultural alignment, and 54% in 3D spatial capabilities. Our results show that our platform offers a feasible approach for future vision-language research and can be applied to various tasks and diverse types of data.
Authors: Chucheng Xiang, Ruchao Bao, Biyin Feng, Wenzheng Wu, Zhongyuan Liu, Yirui Guan, Ligang Liu
Abstract: We present a novel framework for automated interior design that combines large language models (LLMs) with grid-based integer programming to jointly optimize room layout and furniture placement. Given a textual prompt, the LLM-driven agent workflow extracts structured design constraints related to room configurations and furniture arrangements. These constraints are encoded into a unified grid-based representation inspired by ``Modulor". Our formulation accounts for key design requirements, including corridor connectivity, room accessibility, spatial exclusivity, and user-specified preferences. To improve computational efficiency, we adopt a coarse-to-fine optimization strategy that begins with a low-resolution grid to solve a simplified problem and guides the solution at the full resolution. Experimental results across diverse scenarios demonstrate that our joint optimization approach significantly outperforms existing two-stage design pipelines in solution quality, and achieves notable computational efficiency through the coarse-to-fine strategy.
Authors: Jingshan Hong, Haigen Hu, Huihuang Zhang, Qianwei Zhou, Zhao Li
Abstract: In supervised learning, traditional image masking faces two key issues: (i) discarded pixels are underutilized, leading to a loss of valuable contextual information; (ii) masking may remove small or critical features, especially in fine-grained tasks. In contrast, masked image modeling (MIM) has demonstrated that masked regions can be reconstructed from partial input, revealing that even incomplete data can exhibit strong contextual consistency with the original image. This highlights the potential of masked regions as sources of semantic diversity. Motivated by this, we revisit the image masking approach, proposing to treat masked content as auxiliary knowledge rather than ignored. Based on this, we propose MaskAnyNet, which combines masking with a relearning mechanism to exploit both visible and masked information. It can be easily extended to any model with an additional branch to jointly learn from the recomposed masked region. This approach leverages the semantic diversity of the masked regions to enrich features and preserve fine-grained details. Experiments on CNN and Transformer backbones show consistent gains across multiple benchmarks. Further analysis confirms that the proposed method improves semantic diversity through the reuse of masked content.
Authors: Jongseong Bae, Junwoo Ha, Jinnyeong Heo, Yeongin Lee, Ha Young Kim
Abstract: Recent camera-based 3D semantic scene completion (SSC) methods have increasingly explored leveraging temporal cues to enrich the features of the current frame. However, while these approaches primarily focus on enhancing in-frame regions, they often struggle to reconstruct critical out-of-frame areas near the sides of the ego-vehicle, although previous frames commonly contain valuable contextual information about these unseen regions. To address this limitation, we propose the Current-Centric Contextual 3D Fusion (C3DFusion) module, which generates hidden region-aware 3D feature geometry by explicitly aligning 3D-lifted point features from both current and historical frames. C3DFusion performs enhanced temporal fusion through two complementary techniques-historical context blurring and current-centric feature densification-which suppress noise from inaccurately warped historical point features by attenuating their scale, and enhance current point features by increasing their volumetric contribution. Simply integrated into standard SSC architectures, C3DFusion demonstrates strong effectiveness, significantly outperforming state-of-the-art methods on the SemanticKITTI and SSCBench-KITTI-360 datasets. Furthermore, it exhibits robust generalization, achieving notable performance gains when applied to other baseline models.
Authors: Fereidoon Zangeneh, Leonard Bruns, Amit Dekel, Alessandro Pieropan, Patric Jensfelt
Abstract: Accurate camera pose estimation from an image observation in a previously mapped environment is commonly done through structure-based methods: by finding correspondences between 2D keypoints on the image and 3D structure points in the map. In order to make this correspondence search tractable in large scenes, existing pipelines either rely on search heuristics, or perform image retrieval to reduce the search space by comparing the current image to a database of past observations. However, these approaches result in elaborate pipelines or storage requirements that grow with the number of past observations. In this work, we propose a new paradigm for making structure-based relocalisation tractable. Instead of relying on image retrieval or search heuristics, we learn a direct mapping from image observations to the visible scene structure in a compact neural network. Given a query image, a forward pass through our novel visible structure retrieval network allows obtaining the subset of 3D structure points in the map that the image views, thus reducing the search space of 2D-3D correspondences. We show that our proposed method enables performing localisation with an accuracy comparable to the state of the art, while requiring lower computational and storage footprint.
Authors: Jialiang Shen, Jiyang Zheng, Yunqi Xue, Huajie Chen, Yu Yao, Hui Kang, Ruiqi Liu, Helin Gong, Yang Yang, Dadong Wang, Tongliang Liu
Abstract: With growing concerns over image authenticity and digital safety, the field of AI-generated image (AIGI) detection has progressed rapidly. Yet, most AIGI detectors still struggle under real-world degradations, particularly motion blur, which frequently occurs in handheld photography, fast motion, and compressed video. Such blur distorts fine textures and suppresses high-frequency artifacts, causing severe performance drops in real-world settings. We address this limitation with a blur-robust AIGI detection framework based on teacher-student knowledge distillation. A high-capacity teacher (DINOv3), trained on clean (i.e., sharp) images, provides stable and semantically rich representations that serve as a reference for learning. By freezing the teacher to maintain its generalization ability, we distill its feature and logit responses from sharp images to a student trained on blurred counterparts, enabling the student to produce consistent representations under motion degradation. Extensive experiments benchmarks show that our method achieves state-of-the-art performance under both motion-blurred and clean conditions, demonstrating improved generalization and real-world applicability. Source codes will be released at: https://github.com/JiaLiangShen/Dino-Detect-for-blur-robust-AIGC-Detection.
URLs: https://github.com/JiaLiangShen/Dino-Detect-for-blur-robust-AIGC-Detection.
Authors: Jing Li, Yifan Wang, Jiafeng Yan, Renlong Zhang, Bin Yang
Abstract: Infrared and visible image fusion aims to integrate complementary multi-modal information into a single fused result. However, existing methods 1) fail to account for the degradation visible images under adverse weather conditions, thereby compromising fusion performance; and 2) rely on fixed network architectures, limiting their adaptability to diverse degradation scenarios. To address these issues, we propose a one-stop degradation-aware image fusion framework for multi-degradation scenarios driven by a large language model (MdaIF). Given the distinct scattering characteristics of different degradation scenarios (e.g., haze, rain, and snow) in atmospheric transmission, a mixture-of-experts (MoE) system is introduced to tackle image fusion across multiple degradation scenarios. To adaptively extract diverse weather-aware degradation knowledge and scene feature representations, collectively referred to as the semantic prior, we employ a pre-trained vision-language model (VLM) in our framework. Guided by the semantic prior, we propose degradation-aware channel attention module (DCAM), which employ degradation prototype decomposition to facilitate multi-modal feature interaction in channel domain. In addition, to achieve effective expert routing, the semantic prior and channel-domain modulated features are utilized to guide the MoE, enabling robust image fusion in complex degradation scenarios. Extensive experiments validate the effectiveness of our MdaIF, demonstrating superior performance over SOTA methods.
Authors: Zheyuan Zhang, Jiwei Zhang, Boyu Zhou, Linzhimeng Duan, Hong Chen
Abstract: Visual Place Recognition (VPR) aims to determine the geographic location of a query image by retrieving its most visually similar counterpart from a geo-tagged reference database. Recently, the emergence of the powerful visual foundation model, DINOv2, trained in a self-supervised manner on massive datasets, has significantly improved VPR performance. This improvement stems from DINOv2's exceptional feature generalization capabilities but is often accompanied by increased model complexity and computational overhead that impede deployment on resource-constrained devices. To address this challenge, we propose $D^{2}$-VPR, a $D$istillation- and $D$eformable-based framework that retains the strong feature extraction capabilities of visual foundation models while significantly reducing model parameters and achieving a more favorable performance-efficiency trade-off. Specifically, first, we employ a two-stage training strategy that integrates knowledge distillation and fine-tuning. Additionally, we introduce a Distillation Recovery Module (DRM) to better align the feature spaces between the teacher and student models, thereby minimizing knowledge transfer losses to the greatest extent possible. Second, we design a Top-Down-attention-based Deformable Aggregator (TDDA) that leverages global semantic features to dynamically and adaptively adjust the Regions of Interest (ROI) used for aggregation, thereby improving adaptability to irregular structures. Extensive experiments demonstrate that our method achieves competitive performance compared to state-of-the-art approaches. Meanwhile, it reduces the parameter count by approximately 64.2% and FLOPs by about 62.6% (compared to CricaVPR).Code is available at https://github.com/tony19980810/D2VPR.
Authors: Yuan Zhou, Litao Hua, Shilong Jin, Wentao Huang, Haoran Duan
Abstract: Keyframe selection has become essential for video understanding with vision-language models (VLMs) due to limited input tokens and the temporal sparsity of relevant information across video frames. Video understanding often relies on effective keyframes that are not only informative but also causally decisive. To this end, we propose Reinforced Causal Search with Information Bottleneck (ReaSon), a framework that formulates keyframe selection as an optimization problem with the help of a novel Causal Information Bottleneck (CIB), which explicitly defines keyframes as those satisfying both predictive sufficiency and causal necessity. Specifically, ReaSon employs a learnable policy network to select keyframes from a visually relevant pool of candidate frames to capture predictive sufficiency, and then assesses causal necessity via counterfactual interventions. Finally, a composite reward aligned with the CIB principle is designed to guide the selection policy through reinforcement learning. Extensive experiments on NExT-QA, EgoSchema, and Video-MME demonstrate that ReaSon consistently outperforms existing state-of-the-art methods under limited-frame settings, validating its effectiveness and generalization ability.
Authors: Zhiguang Lu, Qianqian Xu, Peisong Wen, Siran Da, Qingming Huang
Abstract: Generative diffusion models show promise for data augmentation. However, applying them to fine-grained tasks presents a significant challenge: ensuring synthetic images accurately capture the subtle, category-defining features critical for high fidelity. Standard approaches, such as text-based Classifier-Free Guidance (CFG), often lack the required specificity, potentially generating misleading examples that degrade fine-grained classifier performance. To address this, we propose Hierarchically Guided Fine-grained Augmentation (HiGFA). HiGFA leverages the temporal dynamics of the diffusion sampling process. It employs strong text and transformed contour guidance with fixed strengths in the early-to-mid sampling stages to establish overall scene, style, and structure. In the final sampling stages, HiGFA activates a specialized fine-grained classifier guidance and dynamically modulates the strength of all guidance signals based on prediction confidence. This hierarchical, confidence-driven orchestration enables HiGFA to generate diverse yet faithful synthetic images by intelligently balancing global structure formation with precise detail refinement. Experiments on several FGVC datasets demonstrate the effectiveness of HiGFA.
Authors: Yijie Guo, Dexiang Hong, Weidong Chen, Zihan She, Cheng Ye, Xiaojun Chang, Zhendong Mao
Abstract: Visual Emotion Analysis (VEA) aims to bridge the affective gap between visual content and human emotional responses. Despite its promise, progress in this field remains limited by the lack of open-source and interpretable datasets. Most existing studies assign a single discrete emotion label to an entire image, offering limited insight into how visual elements contribute to emotion. In this work, we introduce EmoVerse, a large-scale open-source dataset that enables interpretable visual emotion analysis through multi-layered, knowledge-graph-inspired annotations. By decomposing emotions into Background-Attribute-Subject (B-A-S) triplets and grounding each element to visual regions, EmoVerse provides word-level and subject-level emotional reasoning. With over 219k images, the dataset further includes dual annotations in Categorical Emotion States (CES) and Dimensional Emotion Space (DES), facilitating unified discrete and continuous emotion representation. A novel multi-stage pipeline ensures high annotation reliability with minimal human effort. Finally, we introduce an interpretable model that maps visual cues into DES representations and provides detailed attribution explanations. Together, the dataset, pipeline, and model form a comprehensive foundation for advancing explainable high-level emotion understanding.
Authors: Qing Cai, Guihao Yan, Fan Zhang, Cheng Zhang, Zhi Liu
Abstract: Ultrasound standard plane recognition is essential for clinical tasks such as disease screening, organ evaluation, and biometric measurement. However, existing methods fail to effectively exploit shallow structural information and struggle to capture fine-grained semantic differences through contrastive samples generated by image augmentations, ultimately resulting in suboptimal recognition of both structural and discriminative details in ultrasound standard planes. To address these issues, we propose SEMC, a novel Structure-Enhanced Mixture-of-Experts Contrastive learning framework that combines structure-aware feature fusion with expert-guided contrastive learning. Specifically, we first introduce a novel Semantic-Structure Fusion Module (SSFM) to exploit multi-scale structural information and enhance the model's ability to perceive fine-grained structural details by effectively aligning shallow and deep features. Then, a novel Mixture-of-Experts Contrastive Recognition Module (MCRM) is designed to perform hierarchical contrastive learning and classification across multi-level features using a mixture-of-experts (MoE) mechanism, further improving class separability and recognition performance. More importantly, we also curate a large-scale and meticulously annotated liver ultrasound dataset containing six standard planes. Extensive experimental results on our in-house dataset and two public datasets demonstrate that SEMC outperforms recent state-of-the-art methods across various metrics.
Authors: Mohamed Youssef, Lukas Brunner, Klaus Rundhammer, Gerald Czech, Oliver Bimber
Abstract: We introduce a novel method for reconstructing surface temperatures through occluding forest vegetation by combining signal processing and machine learning. Our goal is to enable fully automated aerial wildfire monitoring using autonomous drones, allowing for the early detection of ground fires before smoke or flames are visible. While synthetic aperture (SA) sensing mitigates occlusion from the canopy and sunlight, it introduces thermal blur that obscures the actual surface temperatures. To address this, we train a visual state space model to recover the subtle thermal signals of partially occluded soil and fire hotspots from this blurred data. A key challenge was the scarcity of real-world training data. We overcome this by integrating a latent diffusion model into a vector quantized to generated a large volume of realistic surface temperature simulations from real wildfire recordings, which we further expanded through temperature augmentation and procedural thermal forest simulation. On simulated data across varied ambient and surface temperatures, forest densities, and sunlight conditions, our method reduced the RMSE by a factor of 2 to 2.5 compared to conventional thermal and uncorrected SA imaging. In field experiments focused on high-temperature hotspots, the improvement was even more significant, with a 12.8-fold RMSE gain over conventional thermal and a 2.6-fold gain over uncorrected SA images. We also demonstrate our model's generalization to other thermal signals, such as human signatures for search and rescue. Since simple thresholding is frequently inadequate for detecting subtle thermal signals, the morphological characteristics are equally essential for accurate classification. Our experiments demonstrated another clear advantage: we reconstructed the complete morphology of fire and human signatures, whereas conventional imaging is defeated by partial occlusion.
Authors: Jiayi Zhu, Yihao Huang, Yue Cao, Xiaojun Jia, Qing Guo, Felix Juefei-Xu, Geguang Pu, Bin Wang
Abstract: Large Visual Language Models (LVLMs) now pose a serious yet overlooked privacy threat, as they can infer a social media user's geolocation directly from shared images, leading to unintended privacy leakage. While adversarial image perturbations provide a potential direction for geo-privacy protection, they require relatively strong distortions to be effective against LVLMs, which noticeably degrade visual quality and diminish an image's value for sharing. To overcome this limitation, we identify typographical attacks as a promising direction for protecting geo-privacy by adding text extension outside the visual content. We further investigate which textual semantics are effective in disrupting geolocation inference and design a two-stage, semantics-aware typographical attack that generates deceptive text to protect user privacy. Extensive experiments across three datasets demonstrate that our approach significantly reduces geolocation prediction accuracy of five state-of-the-art commercial LVLMs, establishing a practical and visually-preserving protection strategy against emerging geo-privacy threats.
Authors: Yukuo Ma, Cong Liu, Junke Wang, Junqi Liu, Haibin Huang, Zuxuan Wu, Chi Zhang, Xuelong Li
Abstract: We present TempoMaster, a novel framework that formulates long video generation as next-frame-rate prediction. Specifically, we first generate a low-frame-rate clip that serves as a coarse blueprint of the entire video sequence, and then progressively increase the frame rate to refine visual details and motion continuity. During generation, TempoMaster employs bidirectional attention within each frame-rate level while performing autoregression across frame rates, thus achieving long-range temporal coherence while enabling efficient and parallel synthesis. Extensive experiments demonstrate that TempoMaster establishes a new state-of-the-art in long video generation, excelling in both visual and temporal quality.
Authors: Zuqi Huang, Mengxin Tian, Huan Liu, Wentao Li, Baobao Liang, Jie Wu, Fang Yan, Zhaoqing Tang, Zhongyu Li
Abstract: Accurate cell counting in immunohistochemistry (IHC) images is critical for quantifying protein expression and aiding cancer diagnosis. However, the task remains challenging due to the chromogen overlap, variable biomarker staining, and diverse cellular morphologies. Regression-based counting methods offer advantages over detection-based ones in handling overlapped cells, yet rarely support end-to-end multi-class counting. Moreover, the potential of foundation models remains largely underexplored in this paradigm. To address these limitations, we propose a rank-aware agglomeration framework that selectively distills knowledge from multiple strong foundation models, leveraging their complementary representations to handle IHC heterogeneity and obtain a compact yet effective student model, CountIHC. Unlike prior task-agnostic agglomeration strategies that either treat all teachers equally or rely on feature similarity, we design a Rank-Aware Teacher Selecting (RATS) strategy that models global-to-local patch rankings to assess each teacher's inherent counting capacity and enable sample-wise teacher selection. For multi-class cell counting, we introduce a fine-tuning stage that reformulates the task as vision-language alignment. Discrete semantic anchors derived from structured text prompts encode both category and quantity information, guiding the regression of class-specific density maps and improving counting for overlapping cells. Extensive experiments demonstrate that CountIHC surpasses state-of-the-art methods across 12 IHC biomarkers and 5 tissue types, while exhibiting high agreement with pathologists' assessments. Its effectiveness on H&E-stained data further confirms the scalability of the proposed method.
Authors: Guoqing Xu, Yiheng Li, Yang Yang
Abstract: Precise modeling of lane topology is essential for autonomous driving, as it directly impacts navigation and control decisions.Existing methods typically represent each lane with a single query and infer topological connectivity based on the similarity between lane queries.However, this kind of design struggles to accurately model complex lane structures, leading to unreliable topology prediction.In this view, we propose a Fine-Grained lane topology reasoning framework (TopoFG).It divides the procedure from bird's-eye-view (BEV) features to topology prediction via fine-grained queries into three phases, i.e., Hierarchical Prior Extractor (HPE), Region-Focused Decoder (RFD), and Robust Boundary-Point Topology Reasoning (RBTR).Specifically, HPE extracts global spatial priors from the BEV mask and local sequential priors from in-lane keypoint sequences to guide subsequent fine-grained query modeling.RFD constructs fine-grained queries by integrating the spatial and sequential priors. It then samples reference points in RoI regions of the mask and applies cross-attention with BEV features to refine the query representations of each lane.RBTR models lane connectivity based on boundary-point query features and further employs a topological denoising strategy to reduce matching ambiguity.By integrating spatial and sequential priors into fine-grained queries and applying a denoising strategy to boundary-point topology reasoning, our method precisely models complex lane structures and delivers trustworthy topology predictions.Extensive experiments on the OpenLane-V2 benchmark demonstrate that TopoFG achieves new state-of-the-art performance, with an OLS of 48.0% on subsetA and 45.4% on subsetB.
Authors: Rongkun Zheng, Lu Qi, Xi Chen, Yi Wang, Kun Wang, Hengshuang Zhao
Abstract: While visual autoregressive modeling (VAR) strategies have shed light on image generation with the autoregressive models, their potential for segmentation, a task that requires precise low-level spatial perception, remains unexplored. Inspired by the multi-scale modeling of classic Mask2Former-based models, we propose Seg-VAR, a novel framework that rethinks segmentation as a conditional autoregressive mask generation problem. This is achieved by replacing the discriminative learning with the latent learning process. Specifically, our method incorporates three core components: (1) an image encoder generating latent priors from input images, (2) a spatial-aware seglat (a latent expression of segmentation mask) encoder that maps segmentation masks into discrete latent tokens using a location-sensitive color mapping to distinguish instances, and (3) a decoder reconstructing masks from these latents. A multi-stage training strategy is introduced: first learning seglat representations via image-seglat joint training, then refining latent transformations, and finally aligning image-encoder-derived latents with seglat distributions. Experiments show Seg-VAR outperforms previous discriminative and generative methods on various segmentation tasks and validation benchmarks. By framing segmentation as a sequential hierarchical prediction task, Seg-VAR opens new avenues for integrating autoregressive reasoning into spatial-aware vision systems. Code will be available at https://github.com/rkzheng99/Seg-VAR.
Authors: Ria Shekhawat, Sushrut Patwardhan, Raghavendra Ramachandra, Praveen Kumar Chandaliya, Kishor P. Upla
Abstract: Face Recognition Systems (FRS) are critical for security but remain vulnerable to morphing attacks, where synthetic images blend biometric features from multiple individuals. We propose a novel Single-Image Morphing Attack Detection (S-MAD) approach using a teacher-student framework, where a CNN-based teacher model refines a ViT-based student model. To improve efficiency, we integrate Low-Rank Adaptation (LoRA) for fine-tuning, reducing computational costs while maintaining high detection accuracy. Extensive experiments are conducted on a morphing dataset built from three publicly available face datasets, incorporating ten different morphing generation algorithms to assess robustness. The proposed method is benchmarked against six state-of-the-art S-MAD techniques, demonstrating superior detection performance and computational efficiency.
Authors: Drishya Karki, Merey Ramazanova, Anthony Cioppa, Silvio Giancola, Bernard Ghanem
Abstract: Group Activity Recognition (GAR) is well studied on the video modality for surveillance and indoor team sports (e.g., volleyball, basketball). Yet, other modalities such as agent positions and trajectories over time, i.e. tracking, remain comparatively under-explored despite being compact, agent-centric signals that explicitly encode spatial interactions. Understanding whether pixel (video) or position (tracking) modalities leads to better group activity recognition is therefore important to drive further research on the topic. However, no standardized benchmark currently exists that aligns broadcast video and tracking data for the same group activities, leading to a lack of apples-to-apples comparison between these modalities for GAR. In this work, we introduce SoccerNet-GAR, a multimodal dataset built from the $64$ matches of the football World Cup 2022. Specifically, the broadcast videos and player tracking modalities for $94{,}285$ group activities are synchronized and annotated with $10$ categories. Furthermore, we define a unified evaluation protocol to benchmark two strong unimodal approaches: (i) a competitive video-based classifiers and (ii) a tracking-based classifiers leveraging graph neural networks. In particular, our novel role-aware graph architecture for tracking-based GAR directly encodes tactical structure through positional edges and temporal attention. Our tracking model achieves $67.2\%$ balanced accuracy compared to $58.1\%$ for the best video baseline, while training $4.25 \times$ faster with $438 \times$ fewer parameters ($197K$ \vs $86.3M$). This study provides new insights into the relative strengths of pixels and positions for group activity recognition. Overall, it highlights the importance of modality choice and role-aware modeling for GAR.
Authors: Ziqiong Liu, Yushun Tang, Junyang Ji, Zhihai He
Abstract: Test-time adaptation (TTA) refers to adjusting the model during the testing phase to cope with changes in sample distribution and enhance the model's adaptability to new environments. In real-world scenarios, models often encounter samples from unseen (out-of-distribution, OOD) categories. Misclassifying these as known (in-distribution, ID) classes not only degrades predictive accuracy but can also impair the adaptation process, leading to further errors on subsequent ID samples. Many existing TTA methods suffer substantial performance drops under such conditions. To address this challenge, we propose a Hierarchical Ladder Network that extracts OOD features from class tokens aggregated across all Transformer layers. OOD detection performance is enhanced by combining the original model prediction with the output of the Hierarchical Ladder Network (HLN) via weighted probability fusion. To improve robustness under domain shift, we further introduce an Attention Affine Network (AAN) that adaptively refines the self-attention mechanism conditioned on the token information to better adapt to domain drift, thereby improving the classification performance of the model on datasets with domain shift. Additionally, a weighted entropy mechanism is employed to dynamically suppress the influence of low-confidence samples during adaptation. Experimental results on benchmark datasets show that our method significantly improves the performance on the most widely used classification datasets.
Authors: Artem Moroz, V\'it Zeman, Martin Mik\v{s}\'ik, Elizaveta Isianova, Miroslav David, Pavel Burget, Varun Burde
Abstract: We introduce a unified, end-to-end framework that seamlessly integrates object detection and pose estimation with a versatile onboarding process. Our pipeline begins with an onboarding stage that generates object representations from either traditional 3D CAD models or, in their absence, by rapidly reconstructing a high-fidelity neural representation (NeRF) from multi-view images. Given a test image, our system first employs the CNOS detector to localize target objects. For each detection, our novel pose estimation module, OPFormer, infers the precise 6D pose. The core of OPFormer is a transformer-based architecture that leverages a foundation model for robust feature extraction. It uniquely learns a comprehensive object representation by jointly encoding multiple template views and enriches these features with explicit 3D geometric priors using Normalized Object Coordinate Space (NOCS). A decoder then establishes robust 2D-3D correspondences to determine the final pose. Evaluated on the challenging BOP benchmarks, our integrated system demonstrates a strong balance between accuracy and efficiency, showcasing its practical applicability in both model-based and model-free scenarios.
Authors: Baber Jan, Aiman H. El-Maleh, Abdul Jabbar Siddiqui, Abdul Bais, Saeed Anwar
Abstract: Camouflaged object detection identifies objects that blend seamlessly with their surroundings through similar colors, textures, and patterns. This task challenges both traditional segmentation methods and modern foundation models, which fail dramatically on camouflaged objects. We identify six fundamental challenges in COD: Intrinsic Similarity, Edge Disruption, Extreme Scale Variation, Environmental Complexities, Contextual Dependencies, and Salient-Camouflaged Object Disambiguation. These challenges frequently co-occur and compound the difficulty of detection, requiring comprehensive architectural solutions. We propose C3Net, which addresses all challenges through a specialized dual-pathway decoder architecture. The Edge Refinement Pathway employs gradient-initialized Edge Enhancement Modules to recover precise boundaries from early features. The Contextual Localization Pathway utilizes our novel Image-based Context Guidance mechanism to achieve intrinsic saliency suppression without external models. An Attentive Fusion Module synergistically combines the two pathways via spatial gating. C3Net achieves state-of-the-art performance with S-measures of 0.898 on COD10K, 0.904 on CAMO, and 0.913 on NC4K, while maintaining efficient processing. C3Net demonstrates that complex, multifaceted detection challenges require architectural innovation, with specialized components working synergistically to achieve comprehensive coverage beyond isolated improvements. Code, model weights, and results are available at https://github.com/Baber-Jan/C3Net.
Authors: Yushe Cao, Dianxi Shi, Xing Fu, Xuechao Zou, Haikuo Peng, Xueqi Li, Chun Yu, Junliang Xing
Abstract: While significant progress has been achieved in multimodal facial generation using semantic masks and textual descriptions, conventional feature fusion approaches often fail to enable effective cross-modal interactions, thereby leading to suboptimal generation outcomes. To address this challenge, we introduce MDiTFace--a customized diffusion transformer framework that employs a unified tokenization strategy to process semantic mask and text inputs, eliminating discrepancies between heterogeneous modality representations. The framework facilitates comprehensive multimodal feature interaction through stacked, newly designed multivariate transformer blocks that process all conditions synchronously. Additionally, we design a novel decoupled attention mechanism by dissociating implicit dependencies between mask tokens and temporal embeddings. This mechanism segregates internal computations into dynamic and static pathways, enabling caching and reuse of features computed in static pathways after initial calculation, thereby reducing additional computational overhead introduced by mask condition by over 94% while maintaining performance. Extensive experiments demonstrate that MDiTFace significantly outperforms other competing methods in terms of both facial fidelity and conditional consistency.
Authors: Xunzhi Xiang, Xingye Tian, Guiyu Zhang, Yabo Chen, Shaofeng Zhang, Xuebo Wang, Xin Tao, Qi Fan
Abstract: Variational autoencoders (VAEs) typically encode images into a compact latent space, reducing computational cost but introducing an optimization dilemma: a higher-dimensional latent space improves reconstruction fidelity but often hampers generative performance. Recent methods attempt to address this dilemma by regularizing high-dimensional latent spaces using external vision foundation models (VFMs). However, it remains unclear how high-dimensional VAE latents affect the optimization of generative models. To our knowledge, our analysis is the first to reveal that redundant high-frequency components in high-dimensional latent spaces hinder the training convergence of diffusion models and, consequently, degrade generation quality. To alleviate this problem, we propose a spectral self-regularization strategy to suppress redundant high-frequency noise while simultaneously preserving reconstruction quality. The resulting Denoising-VAE, a ViT-based autoencoder that does not rely on VFMs, produces cleaner, lower-noise latents, leading to improved generative quality and faster optimization convergence. We further introduce a spectral alignment strategy to facilitate the optimization of Denoising-VAE-based generative models. Our complete method enables diffusion models to converge approximately 2$\times$ faster than with SD-VAE, while achieving state-of-the-art reconstruction quality (rFID = 0.28, PSNR = 27.26) and competitive generation performance (gFID = 1.82) on the ImageNet 256$\times$256 benchmark.
Authors: Ye Du, Nanxi Yu, Shujun Wang
Abstract: Vision-language foundation models (VLMs) have shown great potential in feature transfer and generalization across a wide spectrum of medical-related downstream tasks. However, fine-tuning these models is resource-intensive due to their large number of parameters. Prompt tuning has emerged as a viable solution to mitigate memory usage and reduce training time while maintaining competitive performance. Nevertheless, the challenge is that existing prompt tuning methods cannot precisely distinguish different kinds of medical concepts, which miss essentially specific disease-related features across various medical imaging modalities in medical image classification tasks. We find that Large Language Models (LLMs), trained on extensive text corpora, are particularly adept at providing this specialized medical knowledge. Motivated by this, we propose incorporating LLMs into the prompt tuning process. Specifically, we introduce the CILMP, Conditional Intervention of Large Language Models for Prompt Tuning, a method that bridges LLMs and VLMs to facilitate the transfer of medical knowledge into VLM prompts. CILMP extracts disease-specific representations from LLMs, intervenes within a low-rank linear subspace, and utilizes them to create disease-specific prompts. Additionally, a conditional mechanism is incorporated to condition the intervention process on each individual medical image, generating instance-adaptive prompts and thus enhancing adaptability. Extensive experiments across diverse medical image datasets demonstrate that CILMP consistently outperforms state-of-the-art prompt tuning methods, demonstrating its effectiveness. Code is available at https://github.com/usr922/cilmp.
Authors: Cheng Liao
Abstract: Deep learning-based Visual SLAM (vSLAM) systems exhibit exceptional geometric reasoning capabilities, yet their prohibitive computational overhead severely restricts deployment on resource-constrained autonomous platforms. This paper presents a hierarchical quantization optimization framework, DPVO-QAT++ (DPVO-QAT++: Heterogeneous QAT and CUDA Kernel Fusion for High-Performance Deep Patch Visual Odometry). Through the synergistic integration of learnable scale parameterization, a heterogeneous precision design for the Visual Odometry (VO) front-end and back-end (front-end floating-point fake quantization with FP16/FP32; back-end full precision), and GPU-native kernel fusion for fake quantization (custom CUDA kernels), our framework significantly reduces memory footprint and increases processing speed while preserving the trajectory accuracy of the original model. On the TartanAir dataset, our framework achieves an average FPS increase of 52.1%, a 29.1% reduction in median latency, and a 64.9% reduction in peak GPU memory reservation, while maintaining trajectory accuracy (ATE) comparable to the original DPVO model across 32 validation sequences. On the EuRoC dataset, it realizes an average FPS increase of 30.1%, a 23.1% reduction in median latency, and a 37.7% reduction in peak GPU memory reservation, maintaining comparable trajectory accuracy (ATE) across 11 validation sequences. Experimental results demonstrate that DPVO-QAT++ effectively bridges the gap between high-precision deep VO and the efficiency requirements for practical deployment, offering a viable engineering paradigm for the application of this technology on real-world embedded platforms. Keywords: Visual Odometry, Heterogeneous Precision Architecture, Quantization-Aware Training, CUDA Kernel Fusion, Scale-Only Training, Deep Patch Visual Odometry, GPU-Native Kernel Fusion.
Authors: Zeqin Yu, Haotao Xie, Jian Zhang, Jiangqun Ni, Wenkan Su, Jiwu Huang
Abstract: Existing Text Image Forgery Localization (T-IFL) methods often suffer from poor generalization due to the limited scale of real-world datasets and the distribution gap caused by synthetic data that fails to capture the complexity of real-world tampering. To tackle this issue, we propose Fourier Series-based Tampering Synthesis (FSTS), a structured and interpretable framework for synthesizing tampered text images. FSTS first collects 16,750 real-world tampering instances from five representative tampering types, using a structured pipeline that records human-performed editing traces via multi-format logs (e.g., video, PSD, and editing logs). By analyzing these collected parameters and identifying recurring behavioral patterns at both individual and population levels, we formulate a hierarchical modeling framework. Specifically, each individual tampering parameter is represented as a compact combination of basis operation-parameter configurations, while the population-level distribution is constructed by aggregating these behaviors. Since this formulation draws inspiration from the Fourier series, it enables an interpretable approximation using basis functions and their learned weights. By sampling from this modeled distribution, FSTS synthesizes diverse and realistic training data that better reflect real-world forgery traces. Extensive experiments across four evaluation protocols demonstrate that models trained with FSTS data achieve significantly improved generalization on real-world datasets. Dataset is available at \href{https://github.com/ZeqinYu/FSTS}{Project Page}.
Authors: Hongbin Huang, Junwei Li, Tianxin Xie, Zhuang Li, Cekai Weng, Yaodong Yang, Yue Luo, Li Liu, Jing Tang, Zhijing Shao, Zeyu Wang
Abstract: High-fidelity digital humans are increasingly used in interactive applications, yet achieving both visual realism and real-time responsiveness remains a major challenge. We present a high-fidelity, real-time conversational digital human system that seamlessly combines a visually realistic 3D avatar, persona-driven expressive speech synthesis, and knowledge-grounded dialogue generation. To support natural and timely interaction, we introduce an asynchronous execution pipeline that coordinates multi-modal components with minimal latency. The system supports advanced features such as wake word detection, emotionally expressive prosody, and highly accurate, context-aware response generation. It leverages novel retrieval-augmented methods, including history augmentation to maintain conversational flow and intent-based routing for efficient knowledge access. Together, these components form an integrated system that enables responsive and believable digital humans, suitable for immersive applications in communication, education, and entertainment.
Authors: Tushar Anand, Advik Sinha, Abhijit Das
Abstract: In this work, we propose an accurate and real-time optical flow and disparity estimation model by fusing pairwise input images in the proposed non-causal selective state space for dense perception tasks. We propose a non-causal Mamba block-based model that is fast and efficient and aptly manages the constraints present in a real-time applications. Our proposed model reduces inference times while maintaining high accuracy and low GPU usage for optical flow and disparity map generation. The results and analysis, and validation in real-life scenario justify that our proposed model can be used for unified real-time and accurate 3D dense perception estimation tasks. The code, along with the models, can be found at https://github.com/vimstereo/DensePerceptNCSSD
Authors: Saar Stern, Ido Sobol, Or Litany
Abstract: The goal of Novel View Synthesis (NVS) is to generate realistic images of a given content from unseen viewpoints. But how can we trust that a generated image truly reflects the intended transformation? Evaluating its reliability remains a major challenge. While recent generative models, particularly diffusion-based approaches, have significantly improved NVS quality, existing evaluation metrics struggle to assess whether a generated image is both realistic and faithful to the source view and intended viewpoint transformation. Standard metrics, such as pixel-wise similarity and distribution-based measures, often mis-rank incorrect results as they fail to capture the nuanced relationship between the source image, viewpoint change, and generated output. We propose a task-aware evaluation framework that leverages features from a strong NVS foundation model, Zero123, combined with a lightweight tuning step to enhance discrimination. Using these features, we introduce two complementary evaluation metrics: a reference-based score, $D_{\text{PRISM}}$, and a reference-free score, $\text{MMD}_{\text{PRISM}}$. Both reliably identify incorrect generations and rank models in agreement with human preference studies, addressing a fundamental gap in NVS evaluation. Our framework provides a principled and practical approach to assessing synthesis quality, paving the way for more reliable progress in novel view synthesis. To further support this goal, we apply our reference-free metric to six NVS methods across three benchmarks: Toys4K, Google Scanned Objects (GSO), and OmniObject3D, where $\text{MMD}_{\text{PRISM}}$ produces a clear and stable ranking, with lower scores consistently indicating stronger models.
Authors: Subin Varghese, Joshua Gao, Asad Ur Rahman, Vedhus Hoskere
Abstract: Deploying embodied agents that can answer questions about their surroundings in realistic real-world settings remains difficult, partly due to the scarcity of benchmarks that faithfully capture practical operating conditions. We propose infrastructure inspection as a compelling domain for open-vocabulary Embodied Question Answering (EQA): it naturally demands multi-scale reasoning, long-range spatial understanding, and complex semantic relationships, while offering unique evaluation advantages via standardized National Bridge Inventory (NBI) condition ratings (0-9), professional inspection reports, and egocentric imagery. We introduce BridgeEQA, a benchmark of 2,200 open-vocabulary question-answer pairs (in the style of OpenEQA) grounded in professional inspection reports across 200 real-world bridge scenes with 47.93 images on average per scene. Questions require synthesizing visual evidence across multiple images and aligning responses with NBI condition ratings. We further propose a new EQA metric Image Citation Relevance to evaluate the ability of a model to cite relevant images. Evaluations of state-of-the-art vision-language models reveal substantial performance gaps under episodic memory EQA settings. To address this, we propose Embodied Memory Visual Reasoning (EMVR), which formulates inspection as sequential navigation over an image-based scene graph: images are nodes, and an agent takes actions to traverse views, compare evidence, and reason within a Markov decision process. EMVR shows strong performance over the baselines. We publicly release both the dataset and code.
Authors: Shuaike Shen, Ke Liu, Jiaqing Xie, Shangde Gao, Chunhua Shen, Ge Liu, Mireia Crispin-Ortuzar, Shangqi Gao
Abstract: Foundation models for medical image segmentation struggle under out-of-distribution (OOD) shifts, often producing fragmented false positives on OOD tumors. We introduce R$^{2}$Seg, a training-free framework for robust OOD tumor segmentation that operates via a two-stage Reason-and-Reject process. First, the Reason step employs an LLM-guided anatomical reasoning planner to localize organ anchors and generate multi-scale ROIs. Second, the Reject step applies two-sample statistical testing to candidates generated by a frozen foundation model (BiomedParse) within these ROIs. This statistical rejection filter retains only candidates significantly different from normal tissue, effectively suppressing false positives. Our framework requires no parameter updates, making it compatible with zero-update test-time augmentation and avoiding catastrophic forgetting. On multi-center and multi-modal tumor segmentation benchmarks, R$^{2}$Seg substantially improves Dice, specificity, and sensitivity over strong baselines and the original foundation models. Code are available at https://github.com/Eurekashen/R2Seg.
Authors: Sushant Gautam, Michael A. Riegler, P{\aa}l Halvorsen
Abstract: Vision-language models (VLMs) enable open-ended visual question answering but remain prone to hallucinations. We present HEDGE, a unified framework for hallucination detection that combines controlled visual perturbations, semantic clustering, and robust uncertainty metrics. HEDGE integrates sampling, distortion synthesis, clustering (entailment- and embedding-based), and metric computation into a reproducible pipeline applicable across multimodal architectures. Evaluations on VQA-RAD and KvasirVQA-x1 with three representative VLMs (LLaVA-Med, Med-Gemma, Qwen2.5-VL) reveal clear architecture- and prompt-dependent trends. Hallucination detectability is highest for unified-fusion models with dense visual tokenization (Qwen2.5-VL) and lowest for architectures with restricted tokenization (Med-Gemma). Embedding-based clustering often yields stronger separation when applied directly to the generated answers, whereas NLI-based clustering remains advantageous for LLaVA-Med and for longer, sentence-level responses. Across configurations, the VASE metric consistently provides the most robust hallucination signal, especially when paired with embedding clustering and a moderate sampling budget (n ~ 10-15). Prompt design also matters: concise, label-style outputs offer clearer semantic structure than syntactically constrained one-sentence responses. By framing hallucination detection as a geometric robustness problem shaped jointly by sampling scale, prompt structure, model architecture, and clustering strategy, HEDGE provides a principled, compute-aware foundation for evaluating multimodal reliability. The hedge-bench PyPI library enables reproducible and extensible benchmarking, with full code and experimental resources available at https://github.com/Simula/HEDGE .
Authors: Mohamed A. Mabrok, Yalda Zafari
Abstract: State Space Models (SSMs), particularly the Mamba architecture, have recently emerged as powerful alternatives to Transformers for sequence modeling, offering linear computational complexity while achieving competitive performance. Yet, despite their effectiveness, understanding how these Vision SSMs process spatial information remains challenging due to the lack of transparent, attention-like mechanisms. To address this gap, we introduce a controllability-based interpretability framework that quantifies how different parts of the input sequence (tokens or patches) influence the internal state dynamics of SSMs. We propose two complementary formulations: a Jacobian-based method applicable to any SSM architecture that measures influence through the full chain of state propagation, and a Gramian-based approach for diagonal SSMs that achieves superior speed through closed-form analytical solutions. Both methods operate in a single forward pass with linear complexity, requiring no architectural modifications or hyperparameter tuning. We validate our framework through experiments on three diverse medical imaging modalities, demonstrating that SSMs naturally implement hierarchical feature refinement from diffuse low-level textures in early layers to focused, clinically meaningful patterns in deeper layers. Our analysis reveals domain-specific controllability signatures aligned with diagnostic criteria, progressive spatial selectivity across the network hierarchy, and the substantial influence of scanning strategies on attention patterns. Beyond medical imaging, we articulate applications spanning computer vision, natural language processing, and cross-domain tasks. Our framework establishes controllability analysis as a unified, foundational interpretability paradigm for SSMs across all domains. Code and analysis tools will be made available upon publication
Authors: Safaeid Hossain Arib, Rabeya Akter, Abdul Monaf Chowdhury, Md Jubair Ahmed Sourov, Md Mehedi Hasan
Abstract: Object counting has achieved remarkable success on visible instances, yet state-of-the-art (SOTA) methods fail under occlusion, a pervasive challenge in real world deployment. This failure stems from a fundamental architectural limitation where backbone networks encode occluding surfaces rather than target objects, thereby corrupting the feature representations required for accurate enumeration. To address this, we present CountOCC, an amodal counting framework that explicitly reconstructs occluded object features through hierarchical multimodal guidance. Rather than accepting degraded encodings, we synthesize complete representations by integrating spatial context from visible fragments with semantic priors from text and visual embeddings, generating class-discriminative features at occluded locations across multiple pyramid levels. We further introduce a visual equivalence objective that enforces consistency in attention space, ensuring that both occluded and unoccluded views of the same scene produce spatially aligned gradient-based attention maps. Together, these complementary mechanisms preserve discriminative properties essential for accurate counting under occlusion. For rigorous evaluation, we establish occlusion-augmented versions of FSC 147 and CARPK spanning both structured and unstructured scenes. CountOCC achieves SOTA performance on FSC 147 with 26.72% and 20.80% MAE reduction over prior baselines under occlusion in validation and test, respectively. CountOCC also demonstrates exceptional generalization by setting new SOTA results on CARPK with 49.89% MAE reduction and on CAPTUREReal with 28.79% MAE reduction, validating robust amodal counting across diverse visual domains. Code will be released soon.
Authors: Kaiser Hamid, Can Cui, Khandakar Ashrafi Akbar, Ziran Wang, Nade Liang
Abstract: Understanding where drivers look and why they shift their attention is essential for autonomous systems that read human intent and justify their actions. Most existing models rely on large-scale gaze datasets to learn these patterns; however, such datasets are labor-intensive to collect and time-consuming to curate. We present FSDAM (Few-Shot Driver Attention Modeling), a framework that achieves joint attention prediction and caption generation with approximately 100 annotated examples, two orders of magnitude fewer than existing approaches. Our approach introduces a dual-pathway architecture where separate modules handle spatial prediction and caption generation while maintaining semantic consistency through cross-modal alignment. Despite minimal supervision, FSDAM achieves competitive performance on attention prediction, generates coherent, and context-aware explanations. The model demonstrates robust zero-shot generalization across multiple driving benchmarks. This work shows that effective attention-conditioned generation is achievable with limited supervision, opening new possibilities for practical deployment of explainable driver attention systems in data-constrained scenarios.
Authors: Ankita Raj, Chetan Arora
Abstract: Open-vocabulary object detectors (OVODs) unify vision and language to detect arbitrary object categories based on text prompts, enabling strong zero-shot generalization to novel concepts. As these models gain traction in high-stakes applications such as robotics, autonomous driving, and surveillance, understanding their security risks becomes crucial. In this work, we conduct the first study of backdoor attacks on OVODs and reveal a new attack surface introduced by prompt tuning. We propose TrAP (Trigger-Aware Prompt tuning), a multi-modal backdoor injection strategy that jointly optimizes prompt parameters in both image and text modalities along with visual triggers. TrAP enables the attacker to implant malicious behavior using lightweight, learnable prompt tokens without retraining the base model weights, thus preserving generalization while embedding a hidden backdoor. We adopt a curriculum-based training strategy that progressively shrinks the trigger size, enabling effective backdoor activation using small trigger patches at inference. Experiments across multiple datasets show that TrAP achieves high attack success rates for both object misclassification and object disappearance attacks, while also improving clean image performance on downstream datasets compared to the zero-shot setting.
Authors: Parsa Esmaeilkhani, Longin Jan Latecki
Abstract: Vision Language Models (VLMs) mix visual tokens and text tokens. A puzzling issue is the fact that visual tokens most related to the query receive little to no attention in the final layers of the LLM module of VLMs from the answer tokens, where all tokens are treated equally, in particular, visual and language tokens in the LLM attention layers. This fact may result in wrong answers to visual questions, as our experimental results confirm. It appears that the standard next-token prediction (NTP) loss provides an insufficient signal for directing attention to visual tokens. We hypothesize that a more direct supervision of the attention of visual tokens to corresponding language tokens in the LLM module of VLMs will lead to improved performance on visual tasks. To demonstrate that this is indeed the case, we propose a novel loss function that directly supervises the attention of visual tokens. It directly grounds the answer language tokens in images by directing their attention to the relevant visual tokens. This is achieved by aligning the attention distribution of visual tokens to ground truth attention maps with KL divergence. The ground truth attention maps are obtained from task geometry in synthetic cases or from standard grounding annotations (e.g., bounding boxes or point annotations) in real images, and are used inside the LLM for attention supervision without requiring new labels. The obtained KL attention loss (KLAL) when combined with NTP encourages VLMs to attend to relevant visual tokens while generating answer tokens. This results in notable improvements across geometric tasks, pointing, and referring expression comprehension on both synthetic and real-world data, as demonstrated by our experiments. We also introduce a new dataset to evaluate the line tracing abilities of VLMs. Surprisingly, even commercial VLMs do not perform well on this task.
Authors: Amirhossein Hassanzadeh, Bartosz Krawczyk, Michael Saunders, Rob Wible, Keith Krause, Dimah Dera, Jan van Aardt
Abstract: Voxelization is an effective approach to reduce the computational cost of processing Light Detection and Ranging (LiDAR) data, yet it results in a loss of fine-scale structural information. This study explores whether low-level voxel content information, specifically target occupancy percentage within a voxel, can be inferred from high-level voxelized LiDAR point cloud data collected from Digital Imaging and remote Sensing Image Generation (DIRSIG) software. In our study, the targets include bark, leaf, soil, and miscellaneous materials. We propose a multi-target regression approach in the context of imbalanced learning using Kernel Point Convolutions (KPConv). Our research leverages cost-sensitive learning to address class imbalance called density-based relevance (DBR). We employ weighted Mean Saquared Erorr (MSE), Focal Regression (FocalR), and regularization to improve the optimization of KPConv. This study performs a sensitivity analysis on the voxel size (0.25 - 2 meters) to evaluate the effect of various grid representations in capturing the nuances of the forest. This sensitivity analysis reveals that larger voxel sizes (e.g., 2 meters) result in lower errors due to reduced variability, while smaller voxel sizes (e.g., 0.25 or 0.5 meter) exhibit higher errors, particularly within the canopy, where variability is greatest. For bark and leaf targets, error values at smaller voxel size datasets (0.25 and 0.5 meter) were significantly higher than those in larger voxel size datasets (2 meters), highlighting the difficulty in accurately estimating within-canopy voxel content at fine resolutions. This suggests that the choice of voxel size is application-dependent. Our work fills the gap in deep imbalance learning models for multi-target regression and simulated datasets for 3D LiDAR point clouds of forests.
Authors: Colton R. Crum, Adam Czajka
Abstract: Integrating human perceptual priors into the training of neural networks has been shown to raise model generalization, serve as an effective regularizer, and align models with human expertise for applications in high-risk domains. Existing approaches to integrate saliency into model training often rely on internal model mechanisms, which recent research suggests may be unreliable. Our insight is that many challenges associated with saliency-guided training stem from the placement of the guidance approaches solely within the image space. Instead, we move away from the image space, use the model's latent space embeddings to steer human guidance during training, and we propose SAGE (Saliency-Guided Contrastive Embeddings): a loss function that integrates human saliency into network training using contrastive embeddings. We apply salient-preserving and saliency-degrading signal augmentations to the input and capture the changes in embeddings and model logits. We guide the model towards salient features and away from non-salient features using a contrastive triplet loss. Additionally, we perform a sanity check on the logit distributions to ensure that the model outputs match the saliency-based augmentations. We demonstrate a boost in classification performance across both open- and closed-set scenarios against SOTA saliency-based methods, showing SAGE's effectiveness across various backbones, and include experiments to suggest its wide generalization across tasks.
Authors: Nicholas Karris, Luke Durell, Javier Flores, Tegan Emerson
Abstract: It can be shown that Stable Diffusion has a permutation-invariance property with respect to the rows of Contrastive Language-Image Pretraining (CLIP) embedding matrices. This inspired the novel observation that these embeddings can naturally be interpreted as point clouds in a Wasserstein space rather than as matrices in a Euclidean space. This perspective opens up new possibilities for understanding the geometry of embedding space. For example, when interpolating between embeddings of two distinct prompts, we propose reframing the interpolation problem as an optimal transport problem. By solving this optimal transport problem, we compute a shortest path (or geodesic) between embeddings that captures a more natural and geometrically smooth transition through the embedding space. This results in smoother and more coherent intermediate (interpolated) images when rendered by the Stable Diffusion generative model. We conduct experiments to investigate this effect, comparing the quality of interpolated images produced using optimal transport to those generated by other standard interpolation methods. The novel optimal transport--based approach presented indeed gives smoother image interpolations, suggesting that viewing the embeddings as point clouds (rather than as matrices) better reflects and leverages the geometry of the embedding space.
Authors: C\u{a}t\u{a}lin-Alexandru R\^ipanu, Andrei-Theodor Hotnog, Giulia-Stefania Imbrea, Dumitru-Clementin Cercel
Abstract: Automatic sign language recognition plays a crucial role in bridging the communication gap between deaf communities and hearing individuals; however, most available datasets focus on American Sign Language. For Romanian Isolated Sign Language Recognition (RoISLR), no large-scale, standardized dataset exists, which limits research progress. In this work, we introduce a new corpus for RoISLR, named RoCoISLR, comprising over 9,000 video samples that span nearly 6,000 standardized glosses from multiple sources. We establish benchmark results by evaluating seven state-of-the-art video recognition models-I3D, SlowFast, Swin Transformer, TimeSformer, Uniformer, VideoMAE, and PoseConv3D-under consistent experimental setups, and compare their performance with that of the widely used WLASL2000 corpus. According to the results, transformer-based architectures outperform convolutional baselines; Swin Transformer achieved a Top-1 accuracy of 34.1%. Our benchmarks highlight the challenges associated with long-tail class distributions in low-resource sign languages, and RoCoISLR provides the initial foundation for systematic RoISLR research.
Authors: Maria Larchenko, Dmitry Guskov, Alexander Lobashev, Georgy Derevyanko
Abstract: Color harmonization adjusts the colors of an inserted object so that it perceptually matches the surrounding image, resulting in a seamless composite. The harmonization problem naturally arises in augmented reality (AR), yet harmonization algorithms are not currently integrated into AR pipelines because real-time solutions are scarce. In this work, we address color harmonization for AR by proposing a lightweight approach that supports on-device inference. For this, we leverage classical optimal transport theory by training a compact encoder to predict the Monge-Kantorovich transport map. We benchmark our MKL-Harmonizer algorithm against state-of-the-art methods and demonstrate that for real composite AR images our method achieves the best aggregated score. We release our dedicated AR dataset of composite images with pixel-accurate masks and data-gathering toolkit to support further data acquisition by researchers.
Authors: Andrew Zhou
Abstract: Accurate segmentation of brain tumors is vital for diagnosis, surgical planning, and treatment monitoring. Deep learning has advanced on benchmarks, but two issues limit clinical use: no uncertainty estimates for errors and no segmentation of healthy brain structures around tumors for surgery. Current methods fail to unify tumor localization with anatomical context and lack confidence scores. This study presents an uncertainty-aware framework augmenting nnUNet with a channel for voxel-wise uncertainty. Trained on BraTS2023, it yields a correlation of 0.750 and RMSD of 0.047 for uncertainty without hurting tumor accuracy. It predicts uncertainty in one pass, with no extra networks or inferences, aiding clinical decisions. For whole-brain context, a unified model combines normal and cancer datasets, achieving a DSC of 0.81 for brain structures and 0.86 for tumor, with robust key-region performance. Combining both innovations gives the first model outputting tumor in natural surroundings plus an overlaid uncertainty map. Visual checks of outputs show uncertainty offers key insights to evaluate predictions and fix errors, helping informed surgical decisions from AI.
Authors: Leena Alghamdi, Muhammad Usman, Hafeez Anwar, Abdul Bais, Saeed Anwar
Abstract: Camouflaged object detection is an emerging and challenging computer vision task that requires identifying and segmenting objects that blend seamlessly into their environments due to high similarity in color, texture, and size. This task is further complicated by low-light conditions, partial occlusion, small object size, intricate background patterns, and multiple objects. While many sophisticated methods have been proposed for this task, current methods still struggle to precisely detect camouflaged objects in complex scenarios, especially with small and multiple objects, indicating room for improvement. We propose a Multi-Scale Recursive Network that extracts multi-scale features via a Pyramid Vision Transformer backbone and combines them via specialized Attention-Based Scale Integration Units, enabling selective feature merging. For more precise object detection, our decoder recursively refines features by incorporating Multi-Granularity Fusion Units. A novel recursive-feedback decoding strategy is developed to enhance global context understanding, helping the model overcome the challenges in this task. By jointly leveraging multi-scale learning and recursive feature optimization, our proposed method achieves performance gains, successfully detecting small and multiple camouflaged objects. Our model achieves state-of-the-art results on two benchmark datasets for camouflaged object detection and ranks second on the remaining two. Our codes, model weights, and results are available at \href{https://github.com/linaagh98/MSRNet}{https://github.com/linaagh98/MSRNet}.
URLs: https://github.com/linaagh98/MSRNet, https://github.com/linaagh98/MSRNet
Authors: Rohit Kundu, Vishal Mohanty, Hao Xiong, Shan Jia, Athula Balachandran, Amit K. Roy-Chowdhury
Abstract: The proliferation of generative AI has led to hyper-realistic synthetic videos, escalating misuse risks and outstripping binary real/fake detectors. We introduce SAGA (Source Attribution of Generative AI videos), the first comprehensive framework to address the urgent need for AI-generated video source attribution at a large scale. Unlike traditional detection, SAGA identifies the specific generative model used. It uniquely provides multi-granular attribution across five levels: authenticity, generation task (e.g., T2V/I2V), model version, development team, and the precise generator, offering far richer forensic insights. Our novel video transformer architecture, leveraging features from a robust vision foundation model, effectively captures spatio-temporal artifacts. Critically, we introduce a data-efficient pretrain-and-attribute strategy, enabling SAGA to achieve state-of-the-art attribution using only 0.5\% of source-labeled data per class, matching fully supervised performance. Furthermore, we propose Temporal Attention Signatures (T-Sigs), a novel interpretability method that visualizes learned temporal differences, offering the first explanation for why different video generators are distinguishable. Extensive experiments on public datasets, including cross-domain scenarios, demonstrate that SAGA sets a new benchmark for synthetic video provenance, providing crucial, interpretable insights for forensic and regulatory applications.
Authors: Ruiqi Yang, Tian Yun, Zihan Wang, Ellie Pavlick
Abstract: Multimodal large language models (LLMs) have made rapid progress in visual understanding, yet their extension from images to videos often reduces to a naive concatenation of frame tokens. In this work, we investigate what video finetuning brings to multimodal LLMs. We propose Visual Chain-of-Thought (vCoT), an explicit reasoning process that generates transitional event descriptions between consecutive frames. Using vCoT, we systematically compare image-only LVLMs with their video-finetuned counterparts, both with and without access to these transitional cues. Our experiments show that vCoT significantly improves the performance of image-only models on long-form video question answering, while yielding only marginal gains for video-finetuned models. This suggests that the latter already capture frame-to-frame transitions implicitly. Moreover, we find that video models transfer this temporal reasoning ability to purely static settings, outperforming image models' baselines on relational visual reasoning tasks.
Authors: Trung Thanh Nguyen, Yasutomo Kawanishi, Vijay John, Takahiro Komamizu, Ichiro Ide
Abstract: The widespread use of multi-sensor systems has increased research in multi-view action recognition. While existing approaches in multi-view setups with fully overlapping sensors benefit from consistent view coverage, partially overlapping settings where actions are visible in only a subset of views remain underexplored. This challenge becomes more severe in real-world scenarios, as many systems provide only limited input modalities and rely on sequence-level annotations instead of dense frame-level labels. In this study, we propose View-aware Cross-modal Knowledge Distillation (ViCoKD), a framework that distills knowledge from a fully supervised multi-modal teacher to a modality- and annotation-limited student. ViCoKD employs a cross-modal adapter with cross-modal attention, allowing the student to exploit multi-modal correlations while operating with incomplete modalities. Moreover, we propose a View-aware Consistency module to address view misalignment, where the same action may appear differently or only partially across viewpoints. It enforces prediction alignment when the action is co-visible across views, guided by human-detection masks and confidence-weighted Jensen-Shannon divergence between their predicted class distributions. Experiments on the real-world MultiSensor-Home dataset show that ViCoKD consistently outperforms competitive distillation methods across multiple backbones and environments, delivering significant gains and surpassing the teacher model under limited conditions.
Authors: Junyi Ma, Wentao Bao, Jingyi Xu, Guanzhong Sun, Yu Zheng, Erhang Zhang, Xieyuanli Chen, Hesheng Wang
Abstract: Analyzing hand-object interaction in egocentric vision facilitates VR/AR applications and human-robot policy transfer. Existing research has mostly focused on modeling the behavior paradigm of interactive actions (i.e., "how to interact"). However, the more challenging and fine-grained problem of capturing the critical moments of contact and separation between the hand and the target object (i.e., "when to interact") is still underexplored, which is crucial for immersive interactive experiences in mixed reality and robotic motion planning. Therefore, we formulate this problem as temporal interaction localization (TIL). Some recent works extract semantic masks as TIL references, but suffer from inaccurate object grounding and cluttered scenarios. Although current temporal action localization (TAL) methods perform well in detecting verb-noun action segments, they rely on category annotations during training and exhibit limited precision in localizing hand-object contact/separation moments. To address these issues, we propose a novel zero-shot approach dubbed EgoLoc to localize hand-object contact and separation timestamps in egocentric videos. EgoLoc introduces hand-dynamics-guided sampling to generate high-quality visual prompts. It exploits the vision-language model to identify contact/separation attributes, localize specific timestamps, and provide closed-loop feedback for further refinement. EgoLoc eliminates the need for object masks and verb-noun taxonomies, leading to generalizable zero-shot implementation. Comprehensive experiments on the public dataset and our novel benchmarks demonstrate that EgoLoc achieves plausible TIL for egocentric videos. It is also validated to effectively facilitate multiple downstream applications in egocentric vision and robotic manipulation tasks. Code and relevant data will be released at https://github.com/IRMVLab/EgoLoc.
Authors: Zihao Lin, Zhenshan Shi, Sasa Zhao, Hanwei Zhu, Lingyu Zhu, Baoliang Chen, Lei Mo
Abstract: Assessing human creativity through visual outputs, such as drawings, plays a critical role in fields including psychology, education, and cognitive science. However, current assessment practices still rely heavily on expert-based subjective scoring, which is both labor-intensive and inherently subjective. In this paper, we propose a data-driven framework for automatic and interpretable creativity assessment from drawings. Motivated by the cognitive understanding that creativity can emerge from both what is drawn (content) and how it is drawn (style), we reinterpret the creativity score as a function of these two complementary dimensions.Specifically, we first augment an existing creativity labeled dataset with additional annotations targeting content categories. Based on the enriched dataset, we further propose a multi-modal, multi-task learning framework that simultaneously predicts creativity scores, categorizes content types, and extracts stylistic features. In particular, we introduce a conditional learning mechanism that enables the model to adapt its visual feature extraction by dynamically tuning it to creativity-relevant signals conditioned on the drawing's stylistic and semantic cues.Experimental results demonstrate that our model achieves state-of-the-art performance compared to existing regression-based approaches and offers interpretable visualizations that align well with human judgments. The code and annotations will be made publicly available at https://github.com/WonderOfU9/CSCA_PRCV_2025
Authors: Kaixin Zhang, Ruiqing Yang, Yuan Zhang, Shan You, Tao Huang
Abstract: Visual Autoregressive (VAR) models enable efficient image generation via next-scale prediction but face escalating computational costs as sequence length grows. Existing static pruning methods degrade performance by permanently removing weights or tokens, disrupting pretrained dependencies. To address this, we propose ActVAR, a dynamic activation framework that introduces dual sparsity across model weights and token sequences to enhance efficiency without sacrificing capacity. ActVAR decomposes feedforward networks (FFNs) into lightweight expert sub-networks and employs a learnable router to dynamically select token-specific expert subsets based on content. Simultaneously, a gated token selector identifies high-update-potential tokens for computation while reconstructing unselected tokens to preserve global context and sequence alignment. Training employs a two-stage knowledge distillation strategy, where the original VAR model supervises the learning of routing and gating policies to align with pretrained knowledge. Experiments on the ImageNet $256\times 256$ benchmark demonstrate that ActVAR achieves up to $21.2\%$ FLOPs reduction with minimal performance degradation.
Authors: Kaixuan Zhang, Minxian Li, Mingwu Ren, Jiankang Deng, Xiatian Zhu
Abstract: High Dynamic Range (HDR) imaging is essential for professional digital media creation, e.g., filmmaking, virtual production, and photorealistic rendering. However, 3D scene reconstruction has primarily focused on Low Dynamic Range (LDR) data, limiting its applicability to professional workflows. Existing approaches that reconstruct HDR scenes from LDR observations rely on multi-exposure fusion or inverse tone-mapping, which increase capture complexity and depend on synthetic supervision. With the recent emergence of cameras that directly capture native HDR data in a single exposure, we present the first method for 3D scene reconstruction that directly models native HDR observations. We propose {\bf Native High dynamic range 3D Gaussian Splatting (NH-3DGS)}, which preserves the full dynamic range throughout the reconstruction pipeline. Our key technical contribution is a novel luminance-chromaticity decomposition of the color representation that enables direct optimization from native HDR camera data. We demonstrate on both synthetic and real multi-view HDR datasets that NH-3DGS significantly outperforms existing methods in reconstruction quality and dynamic range preservation, enabling professional-grade 3D reconstruction directly from native HDR captures. Code and datasets will be made available.
Authors: Hao Li, Zhenfeng Zhuang, Jingyu Lin, Yu Liu, Yifei Chen, Qiong Peng, Lequan Yu, Liansheng Wang
Abstract: Due to the diversity of brain anatomy and the scarcity of annotated data, supervised anomaly detection for brain MRI remains challenging, driving the development of unsupervised anomaly detection (UAD) approaches. Current UAD methods typically utilize artificially generated noise perturbations on healthy MRIs to train generative models for normal anatomy reconstruction, enabling anomaly detection via residual mapping. However, such simulated anomalies lack the biophysical fidelity and morphological complexity characteristic of true clinical lesions. To advance UAD in brain MRI, we conduct the first systematic frequency-domain analysis of pathological signatures, revealing two key properties: (1) anomalies exhibit unique frequency patterns distinguishable from normal anatomy, and (2) low-frequency signals maintain consistent representations across healthy scans. These insights motivate our Frequency-Decomposition Preprocessing (FDP) framework, the first UAD method to leverage frequency-domain reconstruction for simultaneous pathology suppression and anatomical preservation. FDP can integrate seamlessly with existing anomaly simulation techniques, consistently enhancing detection performance across diverse architectures while maintaining diagnostic fidelity. Experimental results demonstrate that FDP consistently improves anomaly detection performance when integrated with existing methods. Notably, FDP achieves a 17.63% increase in DICE score with LDM while maintaining robust improvements across multiple baselines. The code is available at https://github.com/ls1rius/MRI_FDP.
Authors: Junbo Zou, Haotian Xia, Zhen Ye, Shengjie Zhang, Christopher Lai, Vicente Ordonez, Weining Shen, Hanjie Chen
Abstract: Sports video understanding presents unique challenges, requiring models to perceive high-speed dynamics, comprehend complex rules, and reason over long temporal contexts. While Multimodal Large Language Models (MLLMs) have shown promise in genral domains, the current state of research in sports remains narrowly focused: existing approaches are either single-sport centric, limited to specific tasks, or rely on training-free paradigms that lack robust, learned reasoning process. To address this gap, we introduce DeepSport, the first end-to-end trained MLLM framework designed for multi-task, multi-sport video understanding. DeepSport shifts the paradigm from passive frame processing to active, iterative reasoning, empowering the model to ``think with videos'' by dynamically interrogating content via a specialized frame-extraction tool. To enable this, we propose a data distillation pipeline that synthesizes high-quality Chain-of-Thought (CoT) trajectories from 10 diverse data source, creating a unified resource of 78k training data. We then employ a two-stage training strategy, Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) with a novel gated tool-use reward, to optimize the model's reasoning process. Extensive experiments on the testing benchmark of 6.7k questions demonstrate that DeepSport achieves state-of-the-art performance, significantly outperforming baselines of both proprietary model and open-source models. Our work establishes a new foundation for domain-specific video reasoning to address the complexities of diverse sports.
Authors: Yaohua Zha, Xue Yuerong, Chunlin Fan, Yuansong Wang, Tao Dai, Ke Chen, Shu-Tao Xia
Abstract: Deep learning-based 3D anomaly detection methods have demonstrated significant potential in industrial manufacturing. However, many approaches are specifically designed for anomaly detection tasks, which limits their generalizability to other 3D understanding tasks. In contrast, self-supervised point cloud models aim for general-purpose representation learning, yet our investigation reveals that these classical models are suboptimal at anomaly detection under the unified fine-tuning paradigm. This motivates us to develop a more generalizable 3D model that can effectively detect anomalies without relying on task-specific designs. Interestingly, we find that using only the curvature of each point as its anomaly score already outperforms several classical self-supervised and dedicated anomaly detection models, highlighting the critical role of curvature in 3D anomaly detection. In this paper, we propose a Curvature-Augmented Self-supervised Learning (CASL) framework based on a reconstruction paradigm. Built upon the classical U-Net architecture, our approach introduces multi-scale curvature prompts to guide the decoder in predicting the spatial coordinates of each point. Without relying on any dedicated anomaly detection mechanisms, it achieves leading detection performance through straightforward anomaly classification fine-tuning. Moreover, the learned representations generalize well to standard 3D understanding tasks such as point cloud classification. The code is available at https://github.com/zyh16143998882/CASL.
Authors: Ruishu Zhu, Sida Huang, Ziheng Jiao, Hongyuan Zhang
Abstract: Multimodal Large Language Models (MLLMs) have played an increasingly important role in multimodal intelligence. However, the existing fine-tuning methods often ignore cross-modal heterogeneity, limiting their full potential. In this work, we propose a novel fine-tuning strategy by injecting beneficial random noise, which outperforms previous methods and even surpasses full fine-tuning, with minimal additional parameters. The proposed Multimodal Noise Generator (MuNG) enables efficient modality fine-tuning by injecting customized noise into the frozen MLLMs. Specifically, we reformulate the reasoning process of MLLMs from a variational inference perspective, upon which we design a multimodal noise generator that dynamically analyzes cross-modal relationships in image-text pairs to generate task-adaptive beneficial noise. Injecting this type of noise into the MLLMs effectively suppresses irrelevant semantic components, leading to significantly improved cross-modal representation alignment and enhanced performance on downstream tasks. Experiments on two mainstream MLLMs, QwenVL and LLaVA, demonstrate that our method surpasses full-parameter fine-tuning and other existing fine-tuning approaches, while requiring adjustments to only about $1\sim2\%$ additional parameters. The relevant code is uploaded in the supplementary.
Authors: Dexin Zuo, Ang Li, Wei Wang, Wenxian Yu, Danping Zou
Abstract: Object 6D pose estimation, a crucial task for robotics and augmented reality applications, becomes particularly challenging when dealing with novel objects whose 3D models are not readily available. To reduce dependency on 3D models, recent studies have explored one-reference-based pose estimation, which requires only a single reference view instead of a complete 3D model. However, existing methods that rely on real-valued coordinate regression suffer from limited global consistency due to the local nature of convolutional architectures and face challenges in symmetric or occluded scenarios owing to a lack of uncertainty modeling. We present CoordAR, a novel autoregressive framework for one-reference 6D pose estimation of unseen objects. CoordAR formulates 3D-3D correspondences between the reference and query views as a map of discrete tokens, which is obtained in an autoregressive and probabilistic manner. To enable accurate correspondence regression, CoordAR introduces 1) a novel coordinate map tokenization that enables probabilistic prediction over discretized 3D space; 2) a modality-decoupled encoding strategy that separately encodes RGB appearance and coordinate cues; and 3) an autoregressive transformer decoder conditioned on both position-aligned query features and the partially generated token sequence. With these novel mechanisms, CoordAR significantly outperforms existing methods on multiple benchmarks and demonstrates strong robustness to symmetry, occlusion, and other challenges in real-world tests.
Authors: Huiqiang Sun, Liao Shen, Zhan Peng, Kun Wang, Size Wu, Yuhang Zang, Tianqi Liu, Zihao Huang, Xingyu Zeng, Zhiguo Cao, Wei Li, Chen Change Loy
Abstract: Cinematic storytelling is profoundly shaped by the artful manipulation of photographic elements such as depth of field and exposure. These effects are crucial in conveying mood and creating aesthetic appeal. However, controlling these effects in generative video models remains highly challenging, as most existing methods are restricted to camera motion control. In this paper, we propose CineCtrl, the first video cinematic editing framework that provides fine control over professional camera parameters (e.g., bokeh, shutter speed). We introduce a decoupled cross-attention mechanism to disentangle camera motion from photographic inputs, allowing fine-grained, independent control without compromising scene consistency. To overcome the shortage of training data, we develop a comprehensive data generation strategy that leverages simulated photographic effects with a dedicated real-world collection pipeline, enabling the construction of a large-scale dataset for robust model training. Extensive experiments demonstrate that our model generates high-fidelity videos with precisely controlled, user-specified photographic camera effects.
Authors: Feng Lv, Haoxuan Feng, Zilu Zhang, Chunlong Xia, Yanfeng Li
Abstract: With the rapid advancement of intelligent transportation systems, text-driven image generation and editing techniques have demonstrated significant potential in providing rich, controllable visual scene data for applications such as traffic monitoring and autonomous driving. However, several challenges remain, including insufficient semantic richness of generated traffic elements, limited camera viewpoints, low visual fidelity of synthesized images, and poor alignment between textual descriptions and generated content. To address these issues, we propose a unified text-driven framework for both image generation and editing, leveraging a controllable mask mechanism to seamlessly integrate the two tasks. Furthermore, we incorporate both vehicle-side and roadside multi-view data to enhance the geometric diversity of traffic scenes. Our training strategy follows a two-stage paradigm: first, we perform conceptual learning using large-scale coarse-grained text-image data; then, we fine-tune with fine-grained descriptive data to enhance text-image alignment and detail quality. Additionally, we introduce a mask-region-weighted loss that dynamically emphasizes small yet critical regions during training, thereby substantially enhancing the generation fidelity of small-scale traffic elements. Extensive experiments demonstrate that our method achieves leading performance in text-based image generation and editing within traffic scenes.
Authors: Dianbing Xi, Guoyuan An, Jingsen Zhu, Zhijian Liu, Yuan Liu, Ruiyuan Zhang, Jiayuan Lu, Rui Wang, Yuchi Huo
Abstract: We propose PFAvatar (Pose-Fusion Avatar), a new method that reconstructs high-quality 3D avatars from ``Outfit of the Day'' (OOTD) photos, which exhibit diverse poses, occlusions, and complex backgrounds. Our method consists of two stages: (1) fine-tuning a pose-aware diffusion model from few-shot OOTD examples and (2) distilling a 3D avatar represented by a neural radiance field (NeRF). In the first stage, unlike previous methods that segment images into assets (e.g., garments, accessories) for 3D assembly, which is prone to inconsistency, we avoid decomposition and directly model the full-body appearance. By integrating a pre-trained ControlNet for pose estimation and a novel Condition Prior Preservation Loss (CPPL), our method enables end-to-end learning of fine details while mitigating language drift in few-shot training. Our method completes personalization in just 5 minutes, achieving a 48$\times$ speed-up compared to previous approaches. In the second stage, we introduce a NeRF-based avatar representation optimized by canonical SMPL-X space sampling and Multi-Resolution 3D-SDS. Compared to mesh-based representations that suffer from resolution-dependent discretization and erroneous occluded geometry, our continuous radiance field can preserve high-frequency textures (e.g., hair) and handle occlusions correctly through transmittance. Experiments demonstrate that PFAvatar outperforms state-of-the-art methods in terms of reconstruction fidelity, detail preservation, and robustness to occlusions/truncations, advancing practical 3D avatar generation from real-world OOTD albums. In addition, the reconstructed 3D avatar supports downstream applications such as virtual try-on, animation, and human video reenactment, further demonstrating the versatility and practical value of our approach.
Authors: Botong Zhao, Qijun Shi, Shujing Lyu, Yue Lu
Abstract: Existing industrial anomaly detection methods mainly determine whether an anomaly is present. However, real-world applications also require discovering and classifying multiple anomaly types. Since industrial anomalies are semantically subtle and current methods do not sufficiently exploit image priors, direct clustering approaches often perform poorly. To address these challenges, we propose ProtoAnomalyNCD, a prototype-learning-based framework for discovering unseen anomaly classes of multiple types that can be integrated with various anomaly detection methods. First, to suppress background clutter, we leverage Grounded SAM with text prompts to localize object regions as priors for the anomaly classification network. Next, because anomalies usually appear as subtle and fine-grained patterns on the product, we introduce an Anomaly-Map-Guided Attention block. Within this block, we design a Region Guidance Factor that helps the attention module distinguish among background, object regions, and anomalous regions. By using both localized product regions and anomaly maps as priors, the module enhances anomalous features while suppressing background noise and preserving normal features for contrastive learning. Finally, under a unified prototype-learning framework, ProtoAnomalyNCD discovers and clusters unseen anomaly classes while simultaneously enabling multi-type anomaly classification. We further extend our method to detect unseen outliers, achieving task-level unification. Our method outperforms state-of-the-art approaches on the MVTec AD, MTD, and Real-IAD datasets.
Authors: Wei Jiang, Jiahao Cui, Yizheng Wu, Zhan Peng, Zhiyu Pan, Zhiguo Cao
Abstract: Reconstructing high dynamic range (HDR) images from low dynamic range (LDR) bursts plays an essential role in the computational photography. Impressive progress has been achieved by learning-based algorithms which require LDR-HDR image pairs. However, these pairs are hard to obtain, which motivates researchers to delve into the problem of annotation-efficient HDR image reconstructing: how to achieve comparable performance with limited HDR ground truths (GTs). This work attempts to address this problem from the view of semi-supervised learning where a teacher model generates pseudo HDR GTs for the LDR samples without GTs and a student model learns from pseudo GTs. Nevertheless, the confirmation bias, i.e., the student may learn from the artifacts in pseudo HDR GTs, presents an impediment. To remove this impediment, an uncertainty-based masking process is proposed to discard unreliable parts of pseudo GTs at both pixel and patch levels, then the trusted areas can be learned from by the student. With this novel masking process, our semi-supervised HDR reconstructing method not only outperforms previous annotation-efficient algorithms, but also achieves comparable performance with up-to-date fully-supervised methods by using only 6.7% HDR GTs.
Authors: Taiye Chen, Zihan Ding, Anjian Li, Christina Zhang, Zeqi Xiao, Yisen Wang, Chi Jin
Abstract: Recent advancements in video generation have demonstrated the potential of using video diffusion models as world models, with autoregressive generation of infinitely long videos through masked conditioning. However, such models, usually with local full attention, lack effective memory compression and retrieval for long-term generation beyond the window size, leading to issues of forgetting and spatiotemporal inconsistencies. To enhance the retention of historical information within a fixed memory budget, we introduce a recurrent neural network (RNN) into the diffusion transformer framework. Specifically, a diffusion model incorporating LSTM with attention achieves comparable performance to state-of-the-art RNN blocks, such as TTT and Mamba2. Moreover, existing diffusion-RNN approaches often suffer from performance degradation due to training-inference gap or the lack of overlap across windows. To address these limitations, we propose a novel Recurrent Autoregressive Diffusion (RAD) framework, which executes frame-wise autoregression for memory update and retrieval, consistently across training and inference time. Experiments on Memory Maze and Minecraft datasets demonstrate the superiority of RAD for long video generation, highlighting the efficiency of LSTM in sequence modeling.
Authors: Chen Ma, Ningfei Wang, Junhao Zheng, Qing Guo, Qian Wang, Qi Alfred Chen, Chao Shen
Abstract: Traffic Sign Recognition (TSR) systems play a critical role in Autonomous Driving (AD) systems, enabling real-time detection of road signs, such as STOP and speed limit signs. While these systems are increasingly integrated into commercial vehicles, recent research has exposed their vulnerability to physical-world adversarial appearance attacks. In such attacks, carefully crafted visual patterns are misinterpreted by TSR models as legitimate traffic signs, while remaining inconspicuous or benign to human observers. However, existing adversarial appearance attacks suffer from notable limitations. Pixel-level perturbation-based methods often lack stealthiness and tend to overfit to specific surrogate models, resulting in poor transferability to real-world TSR systems. On the other hand, text-to-image (T2I) diffusion model-based approaches demonstrate limited effectiveness and poor generalization to out-of-distribution sign types. In this paper, we present DiffSign, a novel T2I-based appearance attack framework designed to generate physically robust, highly effective, transferable, practical, and stealthy appearance attacks against TSR systems. To overcome the limitations of prior approaches, we propose a carefully designed attack pipeline that integrates CLIP-based loss and masked prompts to improve attack focus and controllability. We also propose two novel style customization methods to guide visual appearance and improve out-of-domain traffic sign attack generalization and attack stealthiness. We conduct extensive evaluations of DiffSign under varied real-world conditions, including different distances, angles, light conditions, and sign categories. Our method achieves an average physical-world attack success rate of 83.3%, leveraging DiffSign's high effectiveness in attack transferability.
Authors: Daniel Cavadia
Abstract: Precise and real-time detection of gastrointestinal polyps during endoscopic procedures is crucial for early diagnosis and prevention of colorectal cancer. This work presents EndoSight AI, a deep learning architecture developed and evaluated independently to enable accurate polyp localization and detailed boundary delineation. Leveraging the publicly available Hyper-Kvasir dataset, the system achieves a mean Average Precision (mAP) of 88.3% for polyp detection and a Dice coefficient of up to 69% for segmentation, alongside real-time inference speeds exceeding 35 frames per second on GPU hardware. The training incorporates clinically relevant performance metrics and a novel thermal-aware procedure to ensure model robustness and efficiency. This integrated AI solution is designed for seamless deployment in endoscopy workflows, promising to advance diagnostic accuracy and clinical decision-making in gastrointestinal healthcare.
Authors: Mehrab Mustafy Rahman, Jayanth Mohan, Tiberiu Sosea, Cornelia Caragea
Abstract: Semi-supervised learning (SSL) has demonstrated high performance in image classification tasks by effectively utilizing both labeled and unlabeled data. However, existing SSL methods often suffer from poor calibration, with models yielding overconfident predictions that misrepresent actual prediction likelihoods. Recently, neural networks trained with {\tt mixup} that linearly interpolates random examples from the training set have shown better calibration in supervised settings. However, calibration of neural models remains under-explored in semi-supervised settings. Although effective in supervised model calibration, random mixup of pseudolabels in SSL presents challenges due to the overconfidence and unreliability of pseudolabels. In this work, we introduce CalibrateMix, a targeted mixup-based approach that aims to improve the calibration of SSL models while maintaining or even improving their classification accuracy. Our method leverages training dynamics of labeled and unlabeled samples to identify ``easy-to-learn'' and ``hard-to-learn'' samples, which in turn are utilized in a targeted mixup of easy and hard samples. Experimental results across several benchmark image datasets show that our method achieves lower expected calibration error (ECE) and superior accuracy compared to existing SSL approaches.
Authors: Ning Han, Zhenyu Ge, Feng Han, Yuhua Sun, Chengqing Li, Jingjing Chen
Abstract: Concept erasure aims to remove harmful, inappropriate, or copyrighted content from text-to-image diffusion models while preserving non-target semantics. However, existing methods either rely on costly fine-tuning or apply coarse semantic separation, often degrading unrelated concepts and lacking adaptability to evolving concept sets. To alleviate this issue, we propose Graph-Guided Online Concept Erasure (GrOCE), a training-free framework that performs precise and adaptive concept removal through graph-based semantic reasoning. GrOCE models concepts and their interrelations as a dynamic semantic graph, enabling principled reasoning over dependencies and fine-grained isolation of undesired content. It comprises three components: (1) Dynamic Topological Graph Construction for incremental graph building, (2) Adaptive Cluster Identification for multi-hop traversal with similarity-decay scoring, and (3) Selective Edge Severing for targeted edge removal while preserving global semantics. Extensive experiments demonstrate that GrOCE achieves state-of-the-art performance on Concept Similarity (CS) and Fr\'echet Inception Distance (FID) metrics, offering efficient, accurate, and stable concept erasure without retraining.
Authors: Ziqiao Weng, Yaoyu Fang, Jiahe Qian, Xinkun Wang, Lee AD Cooper, Weidong Cai, Bo Zhou
Abstract: Spatial transcriptomics (ST) bridges gene expression and tissue morphology but faces clinical adoption barriers due to technical complexity and prohibitive costs. While computational methods predict gene expression from H&E-stained whole-slide images (WSIs), existing approaches often fail to capture the intricate biological heterogeneity within spots and are susceptible to morphological noise when integrating contextual information from surrounding tissue. To overcome these limitations, we propose HiFusion, a novel deep learning framework that integrates two complementary components. First, we introduce the Hierarchical Intra-Spot Modeling module that extracts fine-grained morphological representations through multi-resolution sub-patch decomposition, guided by a feature alignment loss to ensure semantic consistency across scales. Concurrently, we present the Context-aware Cross-scale Fusion module, which employs cross-attention to selectively incorporate biologically relevant regional context, thereby enhancing representational capacity. This architecture enables comprehensive modeling of both cellular-level features and tissue microenvironmental cues, which are essential for accurate gene expression prediction. Extensive experiments on two benchmark ST datasets demonstrate that HiFusion achieves state-of-the-art performance across both 2D slide-wise cross-validation and more challenging 3D sample-specific scenarios. These results underscore HiFusion's potential as a robust, accurate, and scalable solution for ST inference from routine histopathology.
Authors: Yoonjae Seo, Ermal Elbasani, Jaehong Lee
Abstract: Most neural network quantization methods apply uniform bit precision across spatial regions, ignoring the heterogeneous structural and textural complexity of visual data. This paper introduces MCAQ-YOLO, a morphological complexity-aware quantization framework for object detection. The framework employs five morphological metrics - fractal dimension, texture entropy, gradient variance, edge density, and contour complexity - to characterize local visual morphology and guide spatially adaptive bit allocation. By correlating these metrics with quantization sensitivity, MCAQ-YOLO dynamically adjusts bit precision according to spatial complexity. In addition, a curriculum-based quantization-aware training scheme progressively increases quantization difficulty to stabilize optimization and accelerate convergence. Experimental results demonstrate a strong correlation between morphological complexity and quantization sensitivity and show that MCAQ-YOLO achieves superior detection accuracy and convergence efficiency compared with uniform quantization. On a safety equipment dataset, MCAQ-YOLO attains 85.6 percent mAP@0.5 with an average of 4.2 bits and a 7.6x compression ratio, yielding 3.5 percentage points higher mAP than uniform 4-bit quantization while introducing only 1.8 ms of additional runtime overhead per image. Cross-dataset validation on COCO and Pascal VOC further confirms consistent performance gains, indicating that morphology-driven spatial quantization can enhance efficiency and robustness for computationally constrained, safety-critical visual recognition tasks.
Authors: Yixuan Yang, Luyang Xie, Zhen Luo, Zixiang Zhao, Mingqi Gao, Feng Zheng
Abstract: Building interactive simulators and scalable robot-learning environments requires a large number of articulated assets. However, most existing 3D assets in simulation are rigid, and manually converting them into articulated objects is extremely labor- and cost-intensive. This raises a natural question: can we automatically identify articulable objects in a scene and convert them into articulated assets directly? In this paper, we present ArtiWorld, a scene-aware pipeline that localizes candidate articulable objects from textual scene descriptions and reconstructs executable URDF models that preserve the original geometry. At the core of this pipeline is Arti4URDF, which leverages 3D point cloud, prior knowledge of a large language model (LLM), and a URDF-oriented prompt design to rapidly convert rigid objects into interactive URDF-based articulated objects while maintaining their 3D shape. We evaluate ArtiWorld at three levels: 3D simulated objects, full 3D simulated scenes, and real-world scan scenes. Across all three settings, our method consistently outperforms existing approaches and achieves state-of-the-art performance, while preserving object geometry and correctly capturing object interactivity to produce usable URDF-based articulated models. This provides a practical path toward building interactive, robot-ready simulation environments directly from existing 3D assets. Code and data will be released.
Authors: Aishwarya Agarwal, Srikrishna Karanam, Vineet Gandhi
Abstract: Contrastive vision-language models (VLMs) such as CLIP achieve strong zero-shot recognition yet remain vulnerable to spurious correlations, particularly background over-reliance. We introduce Cluster-based Concept Importance (CCI), a novel interpretability method that uses CLIP's own patch embeddings to group spatial patches into semantically coherent clusters, mask them, and evaluate relative changes in model predictions. CCI sets a new state of the art on faithfulness benchmarks, surpassing prior methods by large margins; for example, it yields more than a twofold improvement on the deletion-AUC metric for MS COCO retrieval. We further propose that CCI, when combined with GroundedSAM, automatically categorizes predictions as foreground- or background-driven, providing a crucial diagnostic ability. Existing benchmarks such as CounterAnimals, however, rely solely on accuracy and implicitly attribute all performance degradation to background correlations. Our analysis shows this assumption to be incomplete, since many errors arise from viewpoint variation, scale shifts, and fine-grained object confusions. To disentangle these effects, we introduce COVAR, a benchmark that systematically varies object foregrounds and backgrounds. Leveraging CCI with COVAR, we present a comprehensive evaluation of eighteen CLIP variants, offering methodological advances and empirical evidence that chart a path toward more robust VLMs.
Authors: Furui Xu, Shaobo Wang, Jiajun Zhang, Chenghao Sun, Haixiang Tang, Linfeng Zhang
Abstract: The growing scale of datasets in deep learning has introduced significant computational challenges. Dataset pruning addresses this challenge by constructing a compact but informative coreset from the full dataset with comparable performance. Previous approaches typically establish scoring metrics based on specific criteria to identify representative samples. However, these methods predominantly rely on sample scores obtained from the model's performance during the training (i.e., fitting) phase. As scoring models achieve near-optimal performance on training data, such fitting-centric approaches induce a dense distribution of sample scores within a narrow numerical range. This concentration reduces the distinction between samples and hinders effective selection. To address this challenge, we conduct dataset pruning from the perspective of generalization, i.e., scoring samples based on models not exposed to them during training. We propose a plug-and-play framework, UNSEEN, which can be integrated into existing dataset pruning methods. Additionally, conventional score-based methods are single-step and rely on models trained solely on the complete dataset, providing limited perspective on the importance of samples. To address this limitation, we scale UNSEEN to multi-step scenarios and propose an incremental selection technique through scoring models trained on varying coresets, and optimize the quality of the coreset dynamically. Extensive experiments demonstrate that our method significantly outperforms existing state-of-the-art (SOTA) methods on CIFAR-10, CIFAR-100, and ImageNet-1K. Notably, on ImageNet-1K, UNSEEN achieves lossless performance while reducing training data by 30\%.
Authors: Lintong Zhang, Kang Yin, Seong-Whan Lee
Abstract: In the domain of non-generative visual counterfactual explanations (CE), traditional techniques frequently involve the substitution of sections within a query image with corresponding sections from distractor images. Such methods have historically overlooked the semantic relevance of the replacement regions to the target object, thereby impairing the model's interpretability and hindering the editing workflow. Addressing these challenges, the present study introduces an innovative methodology named as Weighted Semantic Map with Auto-adaptive Candidate Editing Network (WSAE-Net). Characterized by two significant advancements: the determination of an weighted semantic map and the auto-adaptive candidate editing sequence. First, the generation of the weighted semantic map is designed to maximize the reduction of non-semantic feature units that need to be computed, thereby optimizing computational efficiency. Second, the auto-adaptive candidate editing sequences are designed to determine the optimal computational order among the feature units to be processed, thereby ensuring the efficient generation of counterfactuals while maintaining the semantic relevance of the replacement feature units to the target object. Through comprehensive experimentation, our methodology demonstrates superior performance, contributing to a more lucid and in-depth understanding of visual counterfactual explanations.
Authors: Zewei Chang, Zheng-Peng Duan, Jianxing Zhang, Chun-Le Guo, Siyu Liu, Hyungju Chun, Hyunhee Park, Zikun Liu, Chongyi Li
Abstract: Image retouching aims to enhance visual quality while aligning with users' personalized aesthetic preferences. To address the challenge of balancing controllability and subjectivity, we propose a unified diffusion-based image retouching framework called PerTouch. Our method supports semantic-level image retouching while maintaining global aesthetics. Using parameter maps containing attribute values in specific semantic regions as input, PerTouch constructs an explicit parameter-to-image mapping for fine-grained image retouching. To improve semantic boundary perception, we introduce semantic replacement and parameter perturbation mechanisms in the training process. To connect natural language instructions with visual control, we develop a VLM-driven agent that can handle both strong and weak user instructions. Equipped with mechanisms of feedback-driven rethinking and scene-aware memory, PerTouch better aligns with user intent and captures long-term preferences. Extensive experiments demonstrate each component's effectiveness and the superior performance of PerTouch in personalized image retouching. Code is available at: https://github.com/Auroral703/PerTouch.
Authors: Pengcheng Shi, Jiawei Chen, Jiaqi Liu, Xinglin Zhang, Tao Chen, Lei Li
Abstract: We introduce Medal S, a medical segmentation foundation model that supports native-resolution spatial and textual prompts within an end-to-end trainable framework. Unlike text-only methods lacking spatial awareness, Medal S achieves channel-wise alignment between volumetric prompts and text embeddings, mitigating inaccuracies from resolution mismatches. By preserving full 3D context, it efficiently processes multiple native-resolution masks in parallel, enhancing multi-class segmentation performance. A lightweight 3D convolutional module enables precise voxel-space refinement guided by both prompt types, supporting up to 243 classes across CT, MRI, PET, ultrasound, and microscopy modalities in the BiomedSegFM dataset. Medal S offers two prompting modes: a text-only mode, where model predictions serve as spatial prompts for self-refinement without human input, and a hybrid mode, incorporating manual annotations for enhanced flexibility. For 24-class segmentation, parallel spatial prompting reduces inference time by more than 90% compared to sequential prompting. We propose dynamic resampling to address target-patch ratio imbalance, extending SAT and nnU-Net for data augmentation. Furthermore, we develop optimized text preprocessing, a two-stage inference strategy, and post-processing techniques to improve memory efficiency, precision, and inference speed. On the five-modality average on the validation set, Medal S outperforms SAT with a DSC of 75.44 (vs. 69.83), NSD of 77.34 (vs. 71.06), F1 of 38.24 (vs. 24.88), and DSC TP of 65.46 (vs. 46.97). Medal S achieves excellent performance by harmonizing spatial precision with semantic textual guidance, demonstrating superior efficiency and accuracy in multi-class medical segmentation tasks compared to sequential prompt-based approaches. Medal S will be publicly available at https://github.com/yinghemedical/Medal-S.
Authors: Jihun Park, Kyoungmin Lee, Jongmin Gim, Hyeonseo Jo, Minseok Oh, Wonhyeok Choi, Kyumin Hwang, Jaeyeul Kim, Minwoo Choi, Sunghoon Im
Abstract: We present Infinite-Story, a training-free framework for consistent text-to-image (T2I) generation tailored for multi-prompt storytelling scenarios. Built upon a scale-wise autoregressive model, our method addresses two key challenges in consistent T2I generation: identity inconsistency and style inconsistency. To overcome these issues, we introduce three complementary techniques: Identity Prompt Replacement, which mitigates context bias in text encoders to align identity attributes across prompts; and a unified attention guidance mechanism comprising Adaptive Style Injection and Synchronized Guidance Adaptation, which jointly enforce global style and identity appearance consistency while preserving prompt fidelity. Unlike prior diffusion-based approaches that require fine-tuning or suffer from slow inference, Infinite-Story operates entirely at test time, delivering high identity and style consistency across diverse prompts. Extensive experiments demonstrate that our method achieves state-of-the-art generation performance, while offering over 6X faster inference (1.72 seconds per image) than the existing fastest consistent T2I models, highlighting its effectiveness and practicality for real-world visual storytelling.
Authors: Wenqian Ye, Di Wang, Guangtao Zheng, Bohan Liu, Aidong Zhang
Abstract: Large vision-language models, such as CLIP, have shown strong zero-shot classification performance by aligning images and text in a shared embedding space. However, CLIP models often develop multimodal spurious biases, which is the undesirable tendency to rely on spurious features. For example, CLIP may infer object types in images based on frequently co-occurring backgrounds rather than the object's core features. This bias significantly impairs the robustness of pre-trained CLIP models on out-of-distribution data, where such cross-modal associations no longer hold. Existing methods for mitigating multimodal spurious bias typically require fine-tuning on downstream data or prior knowledge of the bias, which undermines the out-of-the-box usability of CLIP. In this paper, we first theoretically analyze the impact of multimodal spurious bias in zero-shot classification. Based on this insight, we propose Spuriousness-Aware Guided Exploration (SAGE), a simple and effective method that mitigates spurious bias through guided prompt selection. SAGE requires no training, fine-tuning, or external annotations. It explores a space of prompt templates and selects the prompts that induce the largest semantic separation between classes, thereby improving worst-group robustness. Extensive experiments on four real-world benchmark datasets and five popular backbone models demonstrate that SAGE consistently improves zero-shot performance and generalization, outperforming previous zero-shot approaches without any external knowledge or model updates.
Authors: Qingsen Ma, Chen Zou, Dianyun Wang, Jia Wang, Liuyu Xiang, Zhaofeng He
Abstract: Under extremely low-light conditions, novel view synthesis (NVS) faces severe degradation in terms of geometry, color consistency, and radiometric stability. Standard 3D Gaussian Splatting (3DGS) pipelines fail when applied directly to underexposed inputs, as independent enhancement across views causes illumination inconsistencies and geometric distortion. To address this, we present DTGS, a unified framework that tightly couples Retinex-inspired illumination decomposition with thermal-guided 3D Gaussian Splatting for illumination-invariant reconstruction. Unlike prior approaches that treat enhancement as a pre-processing step, DTGS performs joint optimization across enhancement, geometry, and thermal supervision through a cyclic enhancement-reconstruction mechanism. A thermal supervisory branch stabilizes both color restoration and geometry learning by dynamically balancing enhancement, structural, and thermal losses. Moreover, a Retinex-based decomposition module embedded within the 3DGS loop provides physically interpretable reflectance-illumination separation, ensuring consistent color and texture across viewpoints. To evaluate our method, we construct RGBT-LOW, a new multi-view low-light thermal dataset capturing severe illumination degradation. Extensive experiments show that DTGS significantly outperforms existing low-light enhancement and 3D reconstruction baselines, achieving superior radiometric consistency, geometric fidelity, and color stability under extreme illumination.
Authors: Guoyi Zhang, Guangsheng Xu, Siyang Chen, Han Wang, Xiaohu Zhang
Abstract: Moving infrared small target detection is a key component of infrared search and tracking systems, yet it remains extremely challenging due to low signal-to-clutter ratios, severe target-background imbalance, and weak discriminative features. Existing deep learning methods primarily focus on spatio-temporal feature aggregation, but their gains are limited, revealing that the fundamental bottleneck lies in ambiguous per-frame feature representations rather than spatio-temporal modeling itself. Motivated by this insight, we propose BP-FPN, a backpropagation-driven feature pyramid architecture that fundamentally rethinks feature learning for small target. BP-FPN introduces Gradient-Isolated Low-Level Shortcut (GILS) to efficiently incorporate fine-grained target details without inducing shortcut learning, and Directional Gradient Regularization (DGR) to enforce hierarchical feature consistency during backpropagation. The design is theoretically grounded, introduces negligible computational overhead, and can be seamlessly integrated into existing frameworks. Extensive experiments on multiple public datasets show that BP-FPN consistently establishes new state-of-the-art performance. To the best of our knowledge, it is the first FPN designed for this task entirely from the backpropagation perspective.
Authors: King-Man Tam, Satoshi Ikehata, Yuta Asano, Zhaoyi An, Rei Kawakami
Abstract: Universal Photometric Stereo is a promising approach for recovering surface normals without strict lighting assumptions. However, it struggles when multi-illumination cues are unreliable, such as under biased lighting or in shadows or self-occluded regions of complex in-the-wild scenes. We propose GeoUniPS, a universal photometric stereo network that integrates synthetic supervision with high-level geometric priors from large-scale 3D reconstruction models pretrained on massive in-the-wild data. Our key insight is that these 3D reconstruction models serve as visual-geometry foundation models, inherently encoding rich geometric knowledge of real scenes. To leverage this, we design a Light-Geometry Dual-Branch Encoder that extracts both multi-illumination cues and geometric priors from the frozen 3D reconstruction model. We also address the limitations of the conventional orthographic projection assumption by introducing the PS-Perp dataset with realistic perspective projection to enable learning of spatially varying view directions. Extensive experiments demonstrate that GeoUniPS delivers state-of-the-arts performance across multiple datasets, both quantitatively and qualitatively, especially in the complex in-the-wild scenes.
Authors: Zheyuan Hu, Chieh-Hsin Lai, Ge Wu, Yuki Mitsufuji, Stefano Ermon
Abstract: MeanFlow (MF) is a diffusion-motivated generative model that enables efficient few-step generation by learning long jumps directly from noise to data. In practice, it is often used as a latent MF by leveraging the pre-trained Stable Diffusion variational autoencoder (SD-VAE) for high-dimensional data modeling. However, MF training remains computationally demanding and is often unstable. During inference, the SD-VAE decoder dominates the generation cost, and MF depends on complex guidance hyperparameters for class-conditional generation. In this work, we develop an efficient training and sampling scheme for MF in the latent space of a Representation Autoencoder (RAE), where a pre-trained vision encoder (e.g., DINO) provides semantically rich latents paired with a lightweight decoder. We observe that naive MF training in the RAE latent space suffers from severe gradient explosion. To stabilize and accelerate training, we adopt Consistency Mid-Training for trajectory-aware initialization and use a two-stage scheme: distillation from a pre-trained flow matching teacher to speed convergence and reduce variance, followed by an optional bootstrapping stage with a one-point velocity estimator to further reduce deviation from the oracle mean flow. This design removes the need for guidance, simplifies training configurations, and reduces computation in both training and sampling. Empirically, our method achieves a 1-step FID of 2.03, outperforming vanilla MF's 3.43, while reducing sampling GFLOPS by 38% and total training cost by 83% on ImageNet 256. We further scale our approach to ImageNet 512, achieving a competitive 1-step FID of 3.23 with the lowest GFLOPS among all baselines. Code is available at https://github.com/sony/mf-rae.
Authors: Yufei Wen, Yuting Zhang, Jingdan Kang, Hao Ren, Weibin Cheng, Jintai Chen, Kaishun Wu
Abstract: Hyperspectral imaging (HSI) holds great potential for healthcare due to its rich spectral information. However, acquiring HSI data remains costly and technically demanding. Hyperspectral image reconstruction offers a practical solution by recovering HSI data from accessible modalities, such as RGB. While general domain datasets are abundant, the scarcity of human HSI data limits progress in medical applications. To tackle this, we propose SpectralAdapt, a semi-supervised domain adaptation (SSDA) framework that bridges the domain gap between general and human-centered HSI datasets. To fully exploit limited labels and abundant unlabeled data, we enhance spectral reasoning by introducing Spectral Density Masking (SDM), which adaptively masks RGB channels based on their spectral complexity, encouraging recovery of informative regions from complementary cues during consistency training. Furthermore, we introduce Spectral Endmember Representation Alignment (SERA), which derives physically interpretable endmembers from valuable labeled pixels and employs them as domain-invariant anchors to guide unlabeled predictions, with momentum updates ensuring adaptability and stability. These components are seamlessly integrated into SpectralAdapt, a spectral prior-guided framework that effectively mitigates domain shift, spectral degradation, and data scarcity in HSI reconstruction. Experiments on benchmark datasets demonstrate consistent improvements in spectral fidelity, cross-domain generalization, and training stability, highlighting the promise of SSDA as an efficient solution for hyperspectral imaging in healthcare.
Authors: Jiaze Li, Hao Yin, Wenhui Tan, Jingyang Chen, Boshen Xu, Yuxun Qu, Yijing Chen, Jianzhong Ju, Zhenbo Luo, Jian Luan
Abstract: Self-reflection mechanisms that rely on purely text-based rethinking processes perform well in most multimodal tasks. However, when directly applied to long-form video understanding scenarios, they exhibit clear limitations. The fundamental reasons for this lie in two points: (1)long-form video understanding involves richer and more dynamic visual input, meaning rethinking only the text information is insufficient and necessitates a further rethinking process specifically targeting visual information; (2) purely text-based reflection mechanisms lack cross-modal interaction capabilities, preventing them from fully integrating visual information during reflection. Motivated by these insights, we propose REVISOR (REflective VIsual Segment Oriented Reasoning), a novel framework for tool-augmented multimodal reflection. REVISOR enables MLLMs to collaboratively construct introspective reflection processes across textual and visual modalities, significantly enhancing their reasoning capability for long-form video understanding. To ensure that REVISOR can learn to accurately review video segments highly relevant to the question during reinforcement learning, we designed the Dual Attribution Decoupled Reward (DADR) mechanism. Integrated into the GRPO training strategy, this mechanism enforces causal alignment between the model's reasoning and the selected video evidence. Notably, the REVISOR framework significantly enhances long-form video understanding capability of MLLMs without requiring supplementary supervised fine-tuning or external models, achieving impressive results on four benchmarks including VideoMME, LongVideoBench, MLVU, and LVBench.
Authors: Weihua Wang, Yubo Cui, Xiangru Lin, Zhiheng Li, Zheng Fang
Abstract: Vision-based 3D Semantic Scene Completion (SSC) has received growing attention due to its potential in autonomous driving. While most existing approaches follow an ego-centric paradigm by aggregating and diffusing features over the entire scene, they often overlook fine-grained object-level details, leading to semantic and geometric ambiguities, especially in complex environments. To address this limitation, we propose Ocean, an object-centric prediction framework that decomposes the scene into individual object instances to enable more accurate semantic occupancy prediction. Specifically, we first employ a lightweight segmentation model, MobileSAM, to extract instance masks from the input image. Then, we introduce a 3D Semantic Group Attention module that leverages linear attention to aggregate object-centric features in 3D space. To handle segmentation errors and missing instances, we further design a Global Similarity-Guided Attention module that leverages segmentation features for global interaction. Finally, we propose an Instance-aware Local Diffusion module that improves instance features through a generative process and subsequently refines the scene representation in the BEV space. Extensive experiments on the SemanticKITTI and SSCBench-KITTI360 benchmarks demonstrate that Ocean achieves state-of-the-art performance, with mIoU scores of 17.40 and 20.28, respectively.
Authors: Sheng Liu, Yuanzhi Liang, Jiepeng Wang, Sidan Du, Chi Zhang, Xuelong Li
Abstract: We present Uni-Inter, a unified framework for human motion generation that supports a wide range of interaction scenarios: including human-human, human-object, and human-scene-within a single, task-agnostic architecture. In contrast to existing methods that rely on task-specific designs and exhibit limited generalization, Uni-Inter introduces the Unified Interactive Volume (UIV), a volumetric representation that encodes heterogeneous interactive entities into a shared spatial field. This enables consistent relational reasoning and compound interaction modeling. Motion generation is formulated as joint-wise probabilistic prediction over the UIV, allowing the model to capture fine-grained spatial dependencies and produce coherent, context-aware behaviors. Experiments across three representative interaction tasks demonstrate that Uni-Inter achieves competitive performance and generalizes well to novel combinations of entities. These results suggest that unified modeling of compound interactions offers a promising direction for scalable motion synthesis in complex environments.
Authors: Dahyun Chung, Donghyun Shin, Yujin Sung, Seunggi Moon, Jinwoo Jeon, Byung-Jun Lee
Abstract: Contrastive Language-Image Pre-training (CLIP) has demonstrated strong generalization across a wide range of visual tasks by leveraging large-scale English-image pairs. However, its extension to low-resource languages remains limited due to the scarcity of high-quality multilingual image-text data. Existing multilingual vision-language models exhibit consistently low retrieval performance in underrepresented languages including Czech, Finnish, Croatian, Hungarian, and Romanian on the Crossmodal-3600 (XM3600) benchmark. To address this, we propose a lightweight and data-efficient framework for multilingual vision-language alignment. Our approach requires no image-text pairs or text-text pairs and freezes both the pretrained image encoder and multilingual text encoder during training. Only a compact 1.7M-parameter projection module is trained, using a contrastive loss over English representations as semantic anchors. This minimal training setup enables robust multilingual alignment even for languages with limited supervision. Extensive evaluation across multiple multilingual retrieval benchmarks confirms the effectiveness of our method, showing significant gains in five underrepresented languages where existing models typically underperform. These findings highlight the effectiveness of our pivot-based, parameter-efficient alignment strategy for inclusive multimodal learning.
Authors: Zhenying Fang, Richang Hong
Abstract: Open-Vocabulary Temporal Action Localization (OV-TAL) aims to recognize and localize instances of any desired action categories in videos without explicitly curating training data for all categories. Existing methods mostly recognize action categories at a single granularity, which degrades the recognition accuracy of both base and novel action categories. To address these issues, we propose a Multi-Grained Category-Aware Network (MGCA-Net) comprising a localizer, an action presence predictor, a conventional classifier, and a coarse-to-fine classifier. Specifically, the localizer localizes category-agnostic action proposals. For these action proposals, the action presence predictor estimates the probability that they belong to an action instance. At the same time, the conventional classifier predicts the probability of each action proposal over base action categories at the snippet granularity. Novel action categories are recognized by the coarse-to-fine classifier, which first identifies action presence at the video granularity. Finally, it assigns each action proposal to one category from the coarse categories at the proposal granularity. Through coarse-to-fine category awareness for novel actions and the conventional classifier's awareness of base actions, multi-grained category awareness is achieved, effectively enhancing localization performance. Comprehensive evaluations on the THUMOS'14 and ActivityNet-1.3 benchmarks demonstrate that our method achieves state-of-the-art performance. Furthermore, our MGCA-Net achieves state-of-the-art results under the Zero-Shot Temporal Action Localization setting.
Authors: Yan Gong, Jianli Lu, Yongsheng Gao, Jie Zhao, Xiaojuan Zhang, Susanto Rahardja
Abstract: Indoor semantic segmentation is fundamental to computer vision and robotics, supporting applications such as autonomous navigation, augmented reality, and smart environments. Although RGB-D fusion leverages complementary appearance and geometric cues, existing methods often depend on computationally intensive cross-attention mechanisms and insufficiently model intra- and inter-modal feature relationships, resulting in imprecise feature alignment and limited discriminative representation. To address these challenges, we propose DiffPixelFormer, a differential pixel-aware Transformer for RGB-D indoor scene segmentation that simultaneously enhances intra-modal representations and models inter-modal interactions. At its core, the Intra-Inter Modal Interaction Block (IIMIB) captures intra-modal long-range dependencies via self-attention and models inter-modal interactions with the Differential-Shared Inter-Modal (DSIM) module to disentangle modality-specific and shared cues, enabling fine-grained, pixel-level cross-modal alignment. Furthermore, a dynamic fusion strategy balances modality contributions and fully exploits RGB-D information according to scene characteristics. Extensive experiments on the SUN RGB-D and NYUDv2 benchmarks demonstrate that DiffPixelFormer-L achieves mIoU scores of 54.28% and 59.95%, outperforming DFormer-L by 1.78% and 2.75%, respectively. Code is available at https://github.com/gongyan1/DiffPixelFormer.
Authors: Bo Fang, Yuxin Song, Qiangqiang Wu, Haoyuan Sun, Wenhao Wu, Antoni B. Chan
Abstract: Complex video reasoning remains a significant challenge for Multimodal Large Language Models (MLLMs), as current R1-based methodologies often prioritize text-centric reasoning derived from text-based and image-based developments. In video tasks, such strategies frequently underutilize rich visual information, leading to potential shortcut learning and increased susceptibility to hallucination. To foster a more robust, visual-centric video understanding, we start by introducing a novel self-supervised reinforcement learning GRPO algorithm (Pretext-GRPO) within the standard R1 pipeline, in which positive rewards are assigned for correctly solving pretext tasks on transformed visual inputs, which makes the model to non-trivially process the visual information. Building on the effectiveness of Pretext-GRPO, we further propose the ViSS-R1 framework, which streamlines and integrates pretext-task-based self-supervised learning directly into the MLLM's R1 post-training paradigm. Instead of relying solely on sparse visual cues, our framework compels models to reason about transformed visual input by simultaneously processing both pretext questions (concerning transformations) and true user queries. This necessitates identifying the applied transformation and reconstructing the original video to formulate accurate final answers. Comprehensive evaluations on six widely-used video reasoning and understanding benchmarks demonstrate the effectiveness and superiority of our Pretext-GRPO and ViSS-R1 for complex video reasoning. Our codes and models will be publicly available.
Authors: Ruixin Liu, Zejian Yuan
Abstract: Monocular 3D lane detection is challenged by aleatoric uncertainty arising from inherent observation noise. Existing methods rely on simplified geometric assumptions, such as independent point predictions or global planar modeling, failing to capture structural variations and aleatoric uncertainty in real-world scenarios. In this paper, we propose MonoUnc, a bird's-eye view (BEV)-free 3D lane detector that explicitly models aleatoric uncertainty informed by local lane structures. Specifically, 3D lanes are projected onto the front-view (FV) space and approximated by parametric curves. Guided by curve predictions, curve-point query embeddings are dynamically generated for lane point predictions in 3D space. Each segment formed by two adjacent points is modeled as a 3D Gaussian, parameterized by the local structure and uncertainty estimations. Accordingly, a novel 3D Gaussian matching loss is designed to constrain these parameters jointly. Experiments on the ONCE-3DLanes and OpenLane datasets demonstrate that MonoUnc outperforms previous state-of-the-art (SoTA) methods across all benchmarks under stricter evaluation criteria. Additionally, we propose two comprehensive evaluation metrics for ONCE-3DLanes, calculating the average and maximum bidirectional Chamfer distances to quantify global and local errors. Codes are released at https://github.com/lrx02/MonoUnc.
Authors: Zhenghua Li, Hang Chen, Zihao Sun, Kai Li, Xiaolin Hu
Abstract: Accurate segmentation of neural structures in Electron Microscopy (EM) images is paramount for neuroscience. However, this task is challenged by intricate morphologies, low signal-to-noise ratios, and scarce annotations, limiting the accuracy and generalization of existing methods. To address these challenges, we seek to leverage the priors learned by visual foundation models on a vast amount of natural images to better tackle this task. Specifically, we propose a novel framework that can effectively transfer knowledge from Segment Anything 2 (SAM2), which is pre-trained on natural images, to the EM domain. We first use SAM2 to extract powerful, general-purpose features. To bridge the domain gap, we introduce a Feature-Guided Attention module that leverages semantic cues from SAM2 to guide a lightweight encoder, the Fine-Grained Encoder (FGE), in focusing on these challenging regions. Finally, a dual-affinity decoder generates both coarse and refined affinity maps. Experimental results demonstrate that our method achieves performance comparable to state-of-the-art (SOTA) approaches with the SAM2 weights frozen. Upon further fine-tuning on EM data, our method significantly outperforms existing SOTA methods. This study validates that transferring representations pre-trained on natural images, when combined with targeted domain-adaptive guidance, can effectively address the specific challenges in neuron segmentation.
Authors: Reeshoon Sayera, Akash Kumar, Sirshapan Mitra, Prudvi Kamtam, Yogesh S Rawat
Abstract: Appearance-based gait recognition have achieved strong performance on controlled datasets, yet systematic evaluation of its robustness to real-world corruptions and silhouette variability remains lacking. We present RobustGait, a framework for fine-grained robustness evaluation of appearance-based gait recognition systems. RobustGait evaluation spans four dimensions: the type of perturbation (digital, environmental, temporal, occlusion), the silhouette extraction method (segmentation and parsing networks), the architectural capacities of gait recognition models, and various deployment scenarios. The benchmark introduces 15 corruption types at 5 severity levels across CASIA-B, CCPG, and SUSTech1K, with in-the-wild validation on MEVID, and evaluates six state-of-the-art gait systems. We came across several exciting insights. First, applying noise at the RGB level better reflects real-world degradation, and reveal how distortions propagate through silhouette extraction to the downstream gait recognition systems. Second, gait accuracy is highly sensitive to silhouette extractor biases, revealing an overlooked source of benchmark bias. Third, robustness is dependent on both the type of perturbation and the architectural design. Finally, we explore robustness-enhancing strategies, showing that noise-aware training and knowledge distillation improve performance and move toward deployment-ready systems.
Authors: Jiacheng Tang, Mingyue Feng, Jiachao Liu, Yaonong Wang, Jian Pu
Abstract: Modular design of planning-oriented autonomous driving has markedly advanced end-to-end systems. However, existing architectures remain constrained by an over-reliance on ego status, hindering generalization and robust scene understanding. We identify the root cause as an inherent design within these architectures that allows ego status to be easily leveraged as a shortcut. Specifically, the premature fusion of ego status in the upstream BEV encoder allows an information flow from this strong prior to dominate the downstream planning module. To address this challenge, we propose AdaptiveAD, an architectural-level solution based on a multi-context fusion strategy. Its core is a dual-branch structure that explicitly decouples scene perception and ego status. One branch performs scene-driven reasoning based on multi-task learning, but with ego status deliberately omitted from the BEV encoder, while the other conducts ego-driven reasoning based solely on the planning task. A scene-aware fusion module then adaptively integrates the complementary decisions from the two branches to form the final planning trajectory. To ensure this decoupling does not compromise multi-task learning, we introduce a path attention mechanism for ego-BEV interaction and add two targeted auxiliary tasks: BEV unidirectional distillation and autoregressive online mapping. Extensive evaluations on the nuScenes dataset demonstrate that AdaptiveAD achieves state-of-the-art open-loop planning performance. Crucially, it significantly mitigates the over-reliance on ego status and exhibits impressive generalization capabilities across diverse scenarios.
Authors: Yehonatan Elisha, Seffi Cohen, Oren Barkan, Noam Koenigstein
Abstract: Saliency maps are widely used for visual explanations in deep learning, but a fundamental lack of consensus persists regarding their intended purpose and alignment with diverse user queries. This ambiguity hinders the effective evaluation and practical utility of explanation methods.We address this gap by introducing the Reference-Frame $\times$ Granularity (RFxG) taxonomy, a principled conceptual framework that organizes saliency explanations along two essential axes:Reference-Frame: Distinguishing between pointwise ("Why this prediction?") and contrastive ("Why this and not an alternative?") explanations.Granularity: Ranging from fine-grained class-level (e.g., "Why Husky?") to coarse-grained group-level (e.g., "Why Dog?") interpretations.Using the RFxG lens, we demonstrate critical limitations in existing evaluation metrics, which overwhelmingly prioritize pointwise faithfulness while neglecting contrastive reasoning and semantic granularity. To systematically assess explanation quality across both RFxG dimensions, we propose four novel faithfulness metrics. Our comprehensive evaluation framework applies these metrics to ten state-of-the-art saliency methods, four model architectures, and three datasets.By advocating a shift toward user-intent-driven evaluation, our work provides both the conceptual foundation and the practical tools necessary to develop visual explanations that are not only faithful to the underlying model behavior but are also meaningfully aligned with the complexity of human understanding and inquiry.
Authors: Doanh C. Bui, Ba Hung Ngo, Hoai Luan Pham, Khang Nguyen, Ma\"i K. Nguyen, Yasuhiko Nakashima
Abstract: Lifelong learning on Whole Slide Images (WSIs) aims to train or fine-tune a unified model sequentially on cancer-related tasks, reducing the resources and effort required for data transfer and processing, especially given the gigabyte-scale size of WSIs. In this paper, we introduce MergeSlide, a simple yet effective framework that treats lifelong learning as a model merging problem by leveraging a vision-language pathology foundation model. When a new task arrives, it is: 1) defined with class-aware prompts, 2) fine-tuned for a few epochs using an MLP-free backbone, and 3) merged into a unified model using an orthogonal continual merging strategy that preserves performance and mitigates catastrophic forgetting. For inference under the class-incremental learning (CLASS-IL) setting, where task identity is unknown, we introduce Task-to-Class Prompt-aligned (TCP) inference. Specifically, TCP first identifies the most relevant task using task-level prompts and then applies the corresponding class-aware prompts to generate predictions. To evaluate MergeSlide, we conduct experiments on a stream of six TCGA datasets. The results show that MergeSlide outperforms both rehearsal-based continual learning and vision-language zero-shot baselines. Code and data are available at https://github.com/caodoanh2001/MergeSlide.
Authors: Yu Zhu, Dan Zeng, Shuiwang Li, Qijun Zhao, Qiaomu Shen, Bo Tang
Abstract: Recent research in Category-Agnostic Pose Estimation (CAPE) has adopted fixed textual keypoint description as semantic prior for two-stage pose matching frameworks. While this paradigm enhances robustness and flexibility by disentangling the dependency of support images, our critical analysis reveals two inherent limitations of static joint embedding: (1) polysemy-induced cross-category ambiguity during the matching process(e.g., the concept "leg" exhibiting divergent visual manifestations across humans and furniture), and (2) insufficient discriminability for fine-grained intra-category variations (e.g., posture and fur discrepancies between a sleeping white cat and a standing black cat). To overcome these challenges, we propose a new framework that innovatively integrates hierarchical cross-modal interaction with dual-stream feature refinement, enhancing the joint embedding with both class-level and instance-specific cues from textual description and specific images. Experiments on the MP-100 dataset demonstrate that, regardless of the network backbone, CapeNext consistently outperforms state-of-the-art CAPE methods by a large margin.
Authors: Seungjae Kim, SeungJoon Lee, MyeongAh Cho
Abstract: Multi-object tracking (MOT) predominantly follows the tracking-by-detection paradigm, where Kalman filters serve as the standard motion predictor due to computational efficiency but inherently fail on non-linear motion patterns. Conversely, recent data-driven motion predictors capture complex non-linear dynamics but suffer from limited domain generalization and computational overhead. Through extensive analysis, we reveal that even in datasets dominated by non-linear motion, Kalman filter outperforms data-driven predictors in up to 34\% of cases, demonstrating that real-world tracking scenarios inherently involve both linear and non-linear patterns. To leverage this complementarity, we propose PlugTrack, a novel framework that adaptively fuses Kalman filter and data-driven motion predictors through multi-perceptive motion understanding. Our approach employs multi-perceptive motion analysis to generate adaptive blending factors. PlugTrack achieves significant performance gains on MOT17/MOT20 and state-of-the-art on DanceTrack without modifying existing motion predictors. To the best of our knowledge, PlugTrack is the first framework to bridge classical and modern motion prediction paradigms through adaptive fusion in MOT.
Authors: Fengzhi Xu, Ziyuan Yang, Mengyu Sun, Joey Tianyi Zhou, Yi Zhang
Abstract: Medical image enhancement is clinically valuable, but existing methods require large-scale datasets to learn complex pixel-level mappings. However, the substantial training and storage costs associated with these datasets hinder their practical deployment. While dataset distillation (DD) can alleviate these burdens, existing methods mainly target high-level tasks, where multiple samples share the same label. This many-to-one mapping allows distilled data to capture shared semantics and achieve information compression. In contrast, low-level tasks involve a many-to-many mapping that requires pixel-level fidelity, making low-level DD an underdetermined problem, as a small distilled dataset cannot fully constrain the dense pixel-level mappings. To address this, we propose the first low-level DD method for medical image enhancement. We first leverage anatomical similarities across patients to construct the shared anatomical prior based on a representative patient, which serves as the initialization for the distilled data of different patients. This prior is then personalized for each patient using a Structure-Preserving Personalized Generation (SPG) module, which integrates patient-specific anatomical information into the distilled dataset while preserving pixel-level fidelity. For different low-level tasks, the distilled data is used to construct task-specific high- and low-quality training pairs. Patient-specific knowledge is injected into the distilled data by aligning the gradients computed from networks trained on the distilled pairs with those from the corresponding patient's raw data. Notably, downstream users cannot access raw patient data. Instead, only a distilled dataset containing abstract training information is shared, which excludes patient-specific details and thus preserves privacy.
Authors: Jiazhen Yan, Ziqiang Li, Fan Wang, Boyu Wang, Zhangjie Fu
Abstract: The rapid progress of generative models such as GANs and diffusion models has led to the widespread proliferation of AI-generated images, raising concerns about misinformation, privacy violations, and trust erosion in digital media. Although large-scale multimodal models like CLIP offer strong transferable representations for detecting synthetic content, fine-tuning them often induces catastrophic forgetting, which degrades pre-trained priors and limits cross-domain generalization. To address this issue, we propose the Distillation-guided Gradient Surgery Network (DGS-Net), a novel framework that preserves transferable pre-trained priors while suppressing task-irrelevant components. Specifically, we introduce a gradient-space decomposition that separates harmful and beneficial descent directions during optimization. By projecting task gradients onto the orthogonal complement of harmful directions and aligning with beneficial ones distilled from a frozen CLIP encoder, DGS-Net achieves unified optimization of prior preservation and irrelevant suppression. Extensive experiments on 50 generative models demonstrate that our method outperforms state-of-the-art approaches by an average margin of 6.6, achieving superior detection performance and generalization across diverse generation techniques.
Authors: Shuaibin Fan, Senming Zhong, Wenchao Yan, Minglong Xue
Abstract: Image dehazing is an important task in the field of computer vision, aiming at restoring clear and detail-rich visual content from haze-affected images. However, when dealing with complex scenes, existing methods often struggle to strike a balance between fine-grained feature representation of inhomogeneous haze distribution and global consistency modeling. Furthermore, to better learn the common degenerate representation of haze in spatial variations, we propose an unsupervised dehaze method for implicit neural degradation representation. Firstly, inspired by the Kolmogorov-Arnold representation theorem, we propose a mechanism combining the channel-independent and channel-dependent mechanisms, which efficiently enhances the ability to learn from nonlinear dependencies. which in turn achieves good visual perception in complex scenes. Moreover, we design an implicit neural representation to model haze degradation as a continuous function to eliminate redundant information and the dependence on explicit feature extraction and physical models. To further learn the implicit representation of the haze features, we also designed a dense residual enhancement module from it to eliminate redundant information. This achieves high-quality image restoration. Experimental results show that our method achieves competitive dehaze performance on various public and real-world datasets. This project code will be available at https://github.com/Fan-pixel/NeDR-Dehaze.
Authors: Zhaocheng Yu, Kui Jiang, Junjun Jiang, Xianming Liu, Guanglu Sun, Yi Xiao
Abstract: Rain significantly degrades the performance of computer vision systems, particularly in applications like autonomous driving and video surveillance. While existing deraining methods have made considerable progress, they often struggle with fidelity of semantic and spatial details. To address these limitations, we propose the Multi-Prior Hierarchical Mamba (MPHM) network for image deraining. This novel architecture synergistically integrates macro-semantic textual priors (CLIP) for task-level semantic guidance and micro-structural visual priors (DINOv2) for scene-aware structural information. To alleviate potential conflicts between heterogeneous priors, we devise a progressive Priors Fusion Injection (PFI) that strategically injects complementary cues at different decoder levels. Meanwhile, we equip the backbone network with an elaborate Hierarchical Mamba Module (HMM) to facilitate robust feature representation, featuring a Fourier-enhanced dual-path design that concurrently addresses global context modeling and local detail recovery. Comprehensive experiments demonstrate MPHM's state-of-the-art performance, achieving a 0.57 dB PSNR gain on the Rain200H dataset while delivering superior generalization on real-world rainy scenarios.
Authors: Hanzhe Liang, Jie Zhou, Can Gao, Bingyang Guo, Jinbao Wang, Linlin Shen
Abstract: 3D anomaly detection (AD) is a crucial task in computer vision, aiming to identify anomalous points or regions from point cloud data. However, existing methods may encounter challenges when handling point clouds with changes in orientation and position because the resulting features may vary significantly. To address this problem, we propose a novel Rotationally Invariant Features (RIF) framework for 3D AD. Firstly, to remove the adverse effect of variations on point cloud data, we develop a Point Coordinate Mapping (PCM) technique, which maps each point into a rotationally invariant space to maintain consistency of representation. Then, to learn robust and discriminative features, we design a lightweight Convolutional Transform Feature Network (CTF-Net) to extract rotationally invariant features for the memory bank. To improve the ability of the feature extractor, we introduce the idea of transfer learning to pre-train the feature extractor with 3D data augmentation. Experimental results show that the proposed method achieves the advanced performance on the Anomaly-ShapeNet dataset, with an average P-AUROC improvement of 17.7\%, and also gains the best performance on the Real3D-AD dataset, with an average P-AUROC improvement of 1.6\%. The strong generalization ability of RIF has been verified by combining it with traditional feature extraction methods on anomaly detection tasks, demonstrating great potential for industrial applications.
Authors: Yuqi Zhang, Guanying Chen, Jiaxing Chen, Chuanyu Fu, Chuan Huang, Shuguang Cui
Abstract: Reconstructing 3D scenes and synthesizing novel views from sparse input views is a highly challenging task. Recent advances in video diffusion models have demonstrated strong temporal reasoning capabilities, making them a promising tool for enhancing reconstruction quality under sparse-view settings. However, existing approaches are primarily designed for modest viewpoint variations, which struggle in capturing fine-grained details in close-up scenarios since input information is severely limited. In this paper, we present a diffusion-based framework, called CloseUpShot, for close-up novel view synthesis from sparse inputs via point-conditioned video diffusion. Specifically, we observe that pixel-warping conditioning suffers from severe sparsity and background leakage in close-up settings. To address this, we propose hierarchical warping and occlusion-aware noise suppression, enhancing the quality and completeness of the conditioning images for the video diffusion model. Furthermore, we introduce global structure guidance, which leverages a dense fused point cloud to provide consistent geometric context to the diffusion process, to compensate for the lack of globally consistent 3D constraints in sparse conditioning inputs. Extensive experiments on multiple datasets demonstrate that our method outperforms existing approaches, especially in close-up novel view synthesis, clearly validating the effectiveness of our design.
Authors: Hao Long, Silin Zhou, Lisi Chen, Shuo Shang
Abstract: Recent learning-based methods have reduced the computational complexity of traditional trajectory similarity computation, but state-of-the-art (SOTA) methods still fail to leverage the comprehensive spectrum of trajectory information for similarity modeling. To tackle this problem, we propose \textbf{RePo}, a novel method that jointly encodes \textbf{Re}gion-wise and \textbf{Po}int-wise features to capture both spatial context and fine-grained moving patterns. For region-wise representation, the GPS trajectories are first mapped to grid sequences, and spatial context are captured by structural features and semantic context enriched by visual features. For point-wise representation, three lightweight expert networks extract local, correlation, and continuous movement patterns from dense GPS sequences. Then, a router network adaptively fuses the learned point-wise features, which are subsequently combined with region-wise features using cross-attention to produce the final trajectory embedding. To train RePo, we adopt a contrastive loss with hard negative samples to provide similarity ranking supervision. Experiment results show that RePo achieves an average accuracy improvement of 22.2\% over SOTA baselines across all evaluation metrics.
Authors: Zonghao Ying, Moyang Chen, Nizhang Li, Zhiqiang Wang, Wenxin Zhang, Quanchen Zou, Zonglei Jing, Aishan Liu, Xianglong Liu
Abstract: Jailbreak attacks can circumvent model safety guardrails and reveal critical blind spots. Prior attacks on text-to-video (T2V) models typically add adversarial perturbations to obviously unsafe prompts, which are often easy to detect and defend. In contrast, we show that benign-looking prompts containing rich, implicit cues can induce T2V models to generate semantically unsafe videos that both violate policy and preserve the original (blocked) intent. To realize this, we propose VEIL, a jailbreak framework that leverages T2V models' cross-modal associative patterns via a modular prompt design. Specifically, our prompts combine three components: neutral scene anchors, which provide the surface-level scene description extracted from the blocked intent to maintain plausibility; latent auditory triggers, textual descriptions of innocuous-sounding audio events (e.g., creaking, muffled noises) that exploit learned audio-visual co-occurrence priors to bias the model toward particular unsafe visual concepts; and stylistic modulators, cinematic directives (e.g., camera framing, atmosphere) that amplify and stabilize the latent trigger's effect. We formalize attack generation as a constrained optimization over the above modular prompt space and solve it with a guided search procedure that balances stealth and effectiveness. Extensive experiments over 7 T2V models demonstrate the efficacy of our attack, achieving a 23 percent improvement in average attack success rate in commercial models.
Authors: Chenyang Li, Wenbing Tang, Yihao Huang, Sinong Simon Zhan, Ming Hu, Xiaojun Jia, Yang Liu
Abstract: Vision-and-Language Navigation (VLN) agents have made remarkable progress, but their robustness remains insufficiently studied. Existing adversarial evaluations often rely on perturbations that manifest as unusual textures rarely encountered in everyday indoor environments. Errors under such contrived conditions have limited practical relevance, as real-world agents are unlikely to encounter such artificial patterns. In this work, we focus on indoor lighting, an intrinsic yet largely overlooked scene attribute that strongly influences navigation. We propose Indoor Lighting-based Adversarial Attack (ILA), a black-box framework that manipulates global illumination to disrupt VLN agents. Motivated by typical household lighting usage, we design two attack modes: Static Indoor Lighting-based Attack (SILA), where the lighting intensity remains constant throughout an episode, and Dynamic Indoor Lighting-based Attack (DILA), where lights are switched on or off at critical moments to induce abrupt illumination changes. We evaluate ILA on two state-of-the-art VLN models across three navigation tasks. Results show that ILA significantly increases failure rates while reducing trajectory efficiency, revealing previously unrecognized vulnerabilities of VLN agents to realistic indoor lighting variations.
Authors: Junjie Yang, Yuhao Yan, Gang Wu, Yuxuan Wang, Ruoyu Liang, Xinjie Jiang, Xiang Wan, Fenglei Fan, Yongquan Zhang, Feiwei Qin, Changmiao Wan
Abstract: As Vision-Language Models (VLMs) increasingly gain traction in medical applications, clinicians are progressively expecting AI systems not only to generate textual diagnoses but also to produce corresponding medical images that integrate seamlessly into authentic clinical workflows. Despite the growing interest, existing medical visual benchmarks present notable limitations. They often rely on ambiguous queries that lack sufficient relevance to image content, oversimplify complex diagnostic reasoning into closed-ended shortcuts, and adopt a text-centric evaluation paradigm that overlooks the importance of image generation capabilities. To address these challenges, we introduce \textsc{MedGEN-Bench}, a comprehensive multimodal benchmark designed to advance medical AI research. MedGEN-Bench comprises 6,422 expert-validated image-text pairs spanning six imaging modalities, 16 clinical tasks, and 28 subtasks. It is structured into three distinct formats: Visual Question Answering, Image Editing, and Contextual Multimodal Generation. What sets MedGEN-Bench apart is its focus on contextually intertwined instructions that necessitate sophisticated cross-modal reasoning and open-ended generative outputs, moving beyond the constraints of multiple-choice formats. To evaluate the performance of existing systems, we employ a novel three-tier assessment framework that integrates pixel-level metrics, semantic text analysis, and expert-guided clinical relevance scoring. Using this framework, we systematically assess 10 compositional frameworks, 3 unified models, and 5 VLMs.
Authors: Longhui Zheng, Qiming Xia, Xiaolu Chen, Zhaoliang Liu, Chenglu Wen
Abstract: 3D object detection is critical for autonomous driving, yet it remains fundamentally challenging to simultaneously maximize computational efficiency and capture long-range spatial dependencies. We observed that Mamba-based models, with their linear state-space design, capture long-range dependencies at lower cost, offering a promising balance between efficiency and accuracy. However, existing methods rely on axis-aligned scanning within a fixed window, inevitably discarding spatial information. To address this problem, we propose WinMamba, a novel Mamba-based 3D feature-encoding backbone composed of stacked WinMamba blocks. To enhance the backbone with robust multi-scale representation, the WinMamba block incorporates a window-scale-adaptive module that compensates voxel features across varying resolutions during sampling. Meanwhile, to obtain rich contextual cues within the linear state space, we equip the WinMamba layer with a learnable positional encoding and a window-shift strategy. Extensive experiments on the KITTI and Waymo datasets demonstrate that WinMamba significantly outperforms the baseline. Ablation studies further validate the individual contributions of the WSF and AWF modules in improving detection accuracy. The code will be made publicly available.
Authors: Cesar Portocarrero Rodriguez, Laura Vandeweyen, Yosuke Yamamoto
Abstract: The American Society of Civil Engineers has graded Americas infrastructure condition as a C, with the road system receiving a dismal D. Roads are vital to regional economic viability, yet their management, maintenance, and repair processes remain inefficient, relying on outdated manual or laser-based inspection methods that are both costly and time-consuming. With the increasing availability of real-time visual data from autonomous vehicles, there is an opportunity to apply computer vision (CV) methods for advanced road monitoring, providing insights to guide infrastructure rehabilitation efforts. This project explores the use of state-of-the-art CV techniques for road distress segmentation. It begins by evaluating synthetic data generated with Generative Adversarial Networks (GANs) to assess its usefulness for model training. The study then applies Convolutional Neural Networks (CNNs) for road distress segmentation and subsequently examines the transformer-based model MaskFormer. Results show that GAN-generated data improves model performance and that MaskFormer outperforms the CNN model in two metrics: mAP50 and IoU.
Authors: Rifen Lin, Alex Jinpeng Wang, Jiawei Mo, Min Li
Abstract: Multimodal pretraining has revolutionized visual understanding, but its impact on video-based person re-identification (ReID) remains underexplored. Existing approaches often rely on video-text pairs, yet suffer from two fundamental limitations: (1) lack of genuine multimodal pretraining, and (2) text poorly captures fine-grained temporal motion-an essential cue for distinguishing identities in video. In this work, we take a bold departure from text-based paradigms by introducing the first skeleton-driven pretraining framework for ReID. To achieve this, we propose Contrastive Skeleton-Image Pretraining for ReID (CSIP-ReID), a novel two-stage method that leverages skeleton sequences as a spatiotemporally informative modality aligned with video frames. In the first stage, we employ contrastive learning to align skeleton and visual features at sequence level. In the second stage, we introduce a dynamic Prototype Fusion Updater (PFU) to refine multimodal identity prototypes, fusing motion and appearance cues. Moreover, we propose a Skeleton Guided Temporal Modeling (SGTM) module that distills temporal cues from skeleton data and integrates them into visual features. Extensive experiments demonstrate that CSIP-ReID achieves new state-of-the-art results on standard video ReID benchmarks (MARS, LS-VID, iLIDS-VID). Moreover, it exhibits strong generalization to skeleton-only ReID tasks (BIWI, IAS), significantly outperforming previous methods. CSIP-ReID pioneers an annotation-free and motion-aware pretraining paradigm for ReID, opening a new frontier in multimodal representation learning.
Authors: Haodong Wang, Tao Zhuo, Xiuwei Zhang, Hanlin Yin, Wencong Wu, Yanning Zhang
Abstract: Achieving pixel-level registration between SAR and optical images remains a challenging task due to their fundamentally different imaging mechanisms and visual characteristics. Although deep learning has achieved great success in many cross-modal tasks, its performance on SAR-Optical registration tasks is still unsatisfactory. Gradient-based information has traditionally played a crucial role in handcrafted descriptors by highlighting structural differences. However, such gradient cues have not been effectively leveraged in deep learning frameworks for SAR-Optical image matching. To address this gap, we propose SOMA, a dense registration framework that integrates structural gradient priors into deep features and refines alignment through a hybrid matching strategy. Specifically, we introduce the Feature Gradient Enhancer (FGE), which embeds multi-scale, multi-directional gradient filters into the feature space using attention and reconstruction mechanisms to boost feature distinctiveness. Furthermore, we propose the Global-Local Affine-Flow Matcher (GLAM), which combines affine transformation and flow-based refinement within a coarse-to-fine architecture to ensure both structural consistency and local accuracy. Experimental results demonstrate that SOMA significantly improves registration precision, increasing the CMR@1px by 12.29% on the SEN1-2 dataset and 18.50% on the GFGE_SO dataset. In addition, SOMA exhibits strong robustness and generalizes well across diverse scenes and resolutions.
Authors: Zahra Tabatabaei, Jon Sporring
Abstract: According to the World Health Organization, breast cancer claimed the lives of approximately 685,000 women in 2020. Early diagnosis and accurate clinical decision making are critical in reducing this global burden. In this study, we propose THIR, a novel Content-Based Medical Image Retrieval (CBMIR) framework that leverages topological data analysis specifically, Betti numbers derived from persistent homology to characterize and retrieve histopathological images based on their intrinsic structural patterns. Unlike conventional deep learning approaches that rely on extensive training, annotated datasets, and powerful GPU resources, THIR operates entirely without supervision. It extracts topological fingerprints directly from RGB histopathological images using cubical persistence, encoding the evolution of loops as compact, interpretable feature vectors. The similarity retrieval is then performed by computing the distances between these topological descriptors, efficiently returning the top-K most relevant matches. Extensive experiments on the BreaKHis dataset demonstrate that THIR outperforms state of the art supervised and unsupervised methods. It processes the entire dataset in under 20 minutes on a standard CPU, offering a fast, scalable, and training free solution for clinical image retrieval.
Authors: Chao Yang, Boqian Zhang, Jinghao Xu, Guang Jiang
Abstract: Diffusion-based methods have shown great promise in single image super-resolution (SISR); however, existing approaches often produce blurred fine details due to insufficient guidance in the high-frequency domain. To address this issue, we propose a High-Frequency Guided Diffusion Network based on Wavelet Decomposition (HDW-SR), which replaces the conventional U-Net backbone in diffusion frameworks. Specifically, we perform diffusion only on the residual map, allowing the network to focus more effectively on high-frequency information restoration. We then introduce wavelet-based downsampling in place of standard CNN downsampling to achieve multi-scale frequency decomposition, enabling sparse cross-attention between the high-frequency subbands of the pre-super-resolved image and the low-frequency subbands of the diffused image for explicit high-frequency guidance. Moreover, a Dynamic Thresholding Block (DTB) is designed to refine high-frequency selection during the sparse attention process. During upsampling, the invertibility of the wavelet transform ensures low-loss feature reconstruction. Experiments on both synthetic and real-world datasets demonstrate that HDW-SR achieves competitive super-resolution performance, excelling particularly in recovering fine-grained image details. The code will be available after acceptance.
Authors: Alec Sargood, Lemuel Puglisi, Elinor Thompson, Mirco Musolesi, Daniel C. Alexander
Abstract: Tractography is the process of inferring the trajectories of white-matter pathways in the brain from diffusion magnetic resonance imaging (dMRI). Local tractography methods, which construct streamlines by following local fiber orientation estimates stepwise through an image, are prone to error accumulation and high false positive rates, particularly on noisy or low-resolution data. In contrast, global methods, which attempt to optimize a collection of streamlines to maximize compatibility with underlying fiber orientation estimates, are computationally expensive. To address these challenges, we introduce GenTract, the first generative model for global tractography. We frame tractography as a generative task, learning a direct mapping from dMRI to complete, anatomically plausible streamlines. We compare both diffusion-based and flow matching paradigms and evaluate GenTract's performance against state-of-the-art baselines. Notably, GenTract achieves precision 2.1x higher than the next-best method, TractOracle. This advantage becomes even more pronounced in challenging low-resolution and noisy settings, where it outperforms the closest competitor by an order of magnitude. By producing tractograms with high precision on research-grade data while also maintaining reliability on imperfect, lower-resolution data, GenTract represents a promising solution for global tractography.
Authors: Diego Ortego, Marlon Rodr\'iguez, Mario Almagro, Kunal Dahiya, David Jim\'enez, Juan C. SanMiguel
Abstract: Foundation models have revolutionized artificial intelligence across numerous domains, yet their transformative potential remains largely untapped in Extreme Multi-label Classification (XMC). Queries in XMC are associated with relevant labels from extremely large label spaces, where it is critical to strike a balance between efficiency and performance. Therefore, many recent approaches efficiently pose XMC as a maximum inner product search between embeddings learned from small encoder-only transformer architectures. In this paper, we address two important aspects in XMC: how to effectively harness larger decoder-only models, and how to exploit visual information while maintaining computational efficiency. We demonstrate that both play a critical role in XMC separately and can be combined for improved performance. We show that a few billion-size decoder can deliver substantial improvements while keeping computational overhead manageable. Furthermore, our Vision-enhanced eXtreme Multi-label Learning framework (ViXML) efficiently integrates foundation vision models by pooling a single embedding per image. This limits computational growth while unlocking multi-modal capabilities. Remarkably, ViXML with small encoders outperforms text-only decoder in most cases, showing that an image is worth billions of parameters. Finally, we present an extension of existing text-only datasets to exploit visual metadata and make them available for future benchmarking. Comprehensive experiments across four public text-only datasets and their corresponding image enhanced versions validate our proposals' effectiveness, surpassing previous state-of-the-art by up to +8.21\% in P@1 on the largest dataset. ViXML's code is available at https://github.com/DiegoOrtego/vixml.
Authors: Haoran Tang, Meng Cao, Ruyang Liu, Xiaoxi Liang, Linglong Li, Ge Li, Xiaodan Liang
Abstract: Recent advances in Multi-modal Large Language Models (MLLMs) have showcased remarkable capabilities in vision-language understanding. However, enabling robust video spatial reasoning-the ability to comprehend object locations, orientations, and inter-object relationships in dynamic 3D scenes-remains a key unsolved challenge. Existing approaches primarily rely on spatially grounded supervised fine-tuning or reinforcement learning, yet we observe that such models often exhibit query-locked reasoning, focusing narrowly on objects explicitly mentioned in the prompt while ignoring critical contextual cues. To address this limitation, we propose Object-Centric 3D Rollout (OCR), a novel strategy that introduces structured perturbations to the 3D geometry of selected objects during training. By degrading object-specific visual cues and projecting the altered geometry into 2D space, OCR compels the model to reason holistically across the entire scene. We further design a rollout-based training pipeline that jointly leverages vanilla and region-noisy videos to optimize spatial reasoning trajectories. Experiments demonstrate state-of-the-art performance: our 3B-parameter model achieves 47.5% accuracy on VSI-Bench, outperforming several 7B baselines. Ablations confirm OCR's superiority over prior rollout strategies (e.g., T-GRPO, NoisyRollout).
Authors: Ying Jiang, Jiayin Lu, Yunuo Chen, Yumeng He, Kui Wu, Yin Yang, Chenfanfu Jiang
Abstract: Painting embodies a unique form of visual storytelling, where the creation process is as significant as the final artwork. Although recent advances in generative models have enabled visually compelling painting synthesis, most existing methods focus solely on final image generation or patch-based process simulation, lacking explicit stroke structure and failing to produce smooth, realistic shading. In this work, we present a differentiable stroke reconstruction framework that unifies painting, stylized texturing, and smudging to faithfully reproduce the human painting-smudging loop. Given an input image, our framework first optimizes single- and dual-color Bezier strokes through a parallel differentiable paint renderer, followed by a style generation module that synthesizes geometry-conditioned textures across diverse painting styles. We further introduce a differentiable smudge operator to enable natural color blending and shading. Coupled with a coarse-to-fine optimization strategy, our method jointly optimizes stroke geometry, color, and texture under geometric and semantic guidance. Extensive experiments on oil, watercolor, ink, and digital paintings demonstrate that our approach produces realistic and expressive stroke reconstructions, smooth tonal transitions, and richly stylized appearances, offering a unified model for expressive digital painting creation. See our project page for more demos: https://yingjiang96.github.io/DiffPaintWebsite/.
Authors: Soyul Lee, Seungmin Baek, Dongbo Min
Abstract: Monocular 3D object detection is a cost-effective solution for applications like autonomous driving and robotics, but remains fundamentally ill-posed due to inherently ambiguous depth cues. Recent DETR-based methods attempt to mitigate this through global attention and auxiliary depth prediction, yet they still struggle with inaccurate depth estimates. Moreover, these methods often overlook instance-level detection difficulty, such as occlusion, distance, and truncation, leading to suboptimal detection performance. We propose MonoDLGD, a novel Difficulty-Aware Label-Guided Denoising framework that adaptively perturbs and reconstructs ground-truth labels based on detection uncertainty. Specifically, MonoDLGD applies stronger perturbations to easier instances and weaker ones into harder cases, and then reconstructs them to effectively provide explicit geometric supervision. By jointly optimizing label reconstruction and 3D object detection, MonoDLGD encourages geometry-aware representation learning and improves robustness to varying levels of object complexity. Extensive experiments on the KITTI benchmark demonstrate that MonoDLGD achieves state-of-the-art performance across all difficulty levels.
Authors: Alberto Gomez, Jorge Oliveira, Ramon Casero, Agis Chartsias
Abstract: Ultrasound (US) machines display images on a built-in monitor, but routine transfer to hospital systems relies on DICOM. We propose a self-supervised pipeline to extract the US image from a photograph of the monitor. This removes the DICOM bottleneck and enables rapid testing and prototyping of new algorithms. In a proof-of-concept study, the rectified images retained enough visual fidelity to classify cardiac views with a balanced accuracy of 0.79 with respect to the native DICOMs.
Authors: Junhee Lee, ChaeBeen Bang, MyoungChul Kim, MyeongAh Cho
Abstract: Weakly-Supervised Video Anomaly Detection aims to identify anomalous events using only video-level labels, balancing annotation efficiency with practical applicability. However, existing methods often oversimplify the anomaly space by treating all abnormal events as a single category, overlooking the diverse semantic and temporal characteristics intrinsic to real-world anomalies. Inspired by how humans perceive anomalies, by jointly interpreting temporal motion patterns and semantic structures underlying different anomaly types, we propose RefineVAD, a novel framework that mimics this dual-process reasoning. Our framework integrates two core modules. The first, Motion-aware Temporal Attention and Recalibration (MoTAR), estimates motion salience and dynamically adjusts temporal focus via shift-based attention and global Transformer-based modeling. The second, Category-Oriented Refinement (CORE), injects soft anomaly category priors into the representation space by aligning segment-level features with learnable category prototypes through cross-attention. By jointly leveraging temporal dynamics and semantic structure, explicitly models both "how" motion evolves and "what" semantic category it resembles. Extensive experiments on WVAD benchmark validate the effectiveness of RefineVAD and highlight the importance of integrating semantic context to guide feature refinement toward anomaly-relevant patterns.
Authors: Yonghui Yu, Jiahang Cai, Xun Wang, Wenwu Yang
Abstract: Existing multi-person video pose estimation methods typically adopt a two-stage pipeline: detecting individuals in each frame, followed by temporal modeling for single-person pose estimation. This design relies on heuristic operations such as detection, RoI cropping, and non-maximum suppression (NMS), limiting both accuracy and efficiency. In this paper, we present a fully end-to-end framework for multi-person 2D pose estimation in videos, effectively eliminating heuristic operations. A key challenge is to associate individuals across frames under complex and overlapping temporal trajectories. To address this, we introduce a novel Pose-Aware Video transformEr Network (PAVE-Net), which features a spatial encoder to model intra-frame relations and a spatiotemporal pose decoder to capture global dependencies across frames. To achieve accurate temporal association, we propose a pose-aware attention mechanism that enables each pose query to selectively aggregate features corresponding to the same individual across consecutive frames.Additionally, we explicitly model spatiotemporal dependencies among pose keypoints to improve accuracy. Notably, our approach is the first end-to-end method for multi-frame 2D human pose estimation.Extensive experiments show that PAVE-Net substantially outperforms prior image-based end-to-end methods, achieving a \textbf{6.0} mAP improvement on PoseTrack2017, and delivers accuracy competitive with state-of-the-art two-stage video-based approaches, while offering significant gains in efficiency.Project page: https://github.com/zgspose/PAVENet
Authors: Yijia Fan, Jusheng Zhang, Kaitong Cai, Jing Yang, Jian Wang, Keze Wang
Abstract: Despite recent advancements in 3D-text cross-modal alignment, existing state-of-the-art methods still struggle to align fine-grained textual semantics with detailed geometric structures, and their alignment performance degrades significantly when scaling to large-scale 3D databases. To overcome this limitation, we introduce 3DAlign-DAER, a unified framework designed to align text and 3D geometry via the proposed dynamic attention policy and the efficient retrieval strategy, capturing subtle correspondences for diverse cross-modal retrieval and classification tasks. Specifically, during the training, our proposed dynamic attention policy (DAP) employs the Hierarchical Attention Fusion (HAF) module to represent the alignment as learnable fine-grained token-to-point attentions. To optimize these attentions across different tasks and geometric hierarchies, our DAP further exploits the Monte Carlo tree search to dynamically calibrate HAF attention weights via a hybrid reward signal and further enhances the alignment between textual descriptions and local 3D geometry. During the inference, our 3DAlign-DAER introduces an Efficient Retrieval Strategy (ERS) to leverage efficient hierarchical searching in the large-scale embedding spaces, outperforming traditional methods (e.g., KNN) in accuracy and efficiency. Furthermore, to facilitate text-3D alignment research and train our 3DAlign-DAER, we construct Align3D-2M, a large-scale dataset featuring 2M text-3D pairs, to provide sufficient fine-grained cross-modal annotations. Extensive and comprehensive experiments demonstrate the superior performance of our 3DAlign-DAER on diverse benchmarks. We will release our codes, models, and datasets.
Authors: Qida Tan, Hongyu Yang, Wenchao Du
Abstract: Appearance-based gaze estimation, aiming to predict accurate 3D gaze direction from a single facial image, has made promising progress in recent years. However, most methods suffer significant performance degradation in cross-domain evaluation due to interference from gaze-irrelevant factors, such as expressions, wearables, and image quality. To alleviate this problem, we present a novel Hybrid-domain Adaptative Representation Learning (shorted by HARL) framework that exploits multi-source hybrid datasets to learn robust gaze representation. More specifically, we propose to disentangle gaze-relevant representation from low-quality facial images by aligning features extracted from high-quality near-eye images in an unsupervised domain-adaptation manner, which hardly requires any computational or inference costs. Additionally, we analyze the effect of head-pose and design a simple yet efficient sparse graph fusion module to explore the geometric constraint between gaze direction and head-pose, leading to a dense and robust gaze representation. Extensive experiments on EyeDiap, MPIIFaceGaze, and Gaze360 datasets demonstrate that our approach achieves state-of-the-art accuracy of $\textbf{5.02}^{\circ}$ and $\textbf{3.36}^{\circ}$, and $\textbf{9.26}^{\circ}$ respectively, and present competitive performances through cross-dataset evaluation. The code is available at https://github.com/da60266/HARL.
Authors: Malek Al Abed, Sebiha Demir, Anne Groteklaes, Elodie Germani, Shahrooz Faghihroohi, Hemmen Sabir, Shadi Albarqouni
Abstract: Portable ultra-low-field MRI (uLF-MRI, 0.064 T) offers accessible neuroimaging for neonatal care but suffers from low signal-to-noise ratio and poor diagnostic quality compared to high-field (HF) MRI. We propose MRIQT, a 3D conditional diffusion framework for image quality transfer (IQT) from uLF to HF MRI. MRIQT combines realistic K-space degradation for physics-consistent uLF simulation, v-prediction with classifier-free guidance for stable image-to-image generation, and an SNR-weighted 3D perceptual loss for anatomical fidelity. The model denoises from a noised uLF input conditioned on the same scan, leveraging volumetric attention-UNet architecture for structure-preserving translation. Trained on a neonatal cohort with diverse pathologies, MRIQT surpasses recent GAN and CNN baselines in PSNR 15.3% with 1.78% over the state of the art, while physicians rated 85% of its outputs as good quality with clear pathology present. MRIQT enables high-fidelity, diffusion-based enhancement of portable ultra-low-field (uLF) MRI for deliable neonatal brain assessment.
Authors: Junjie Wu, Guohong Fu
Abstract: Multimodal misinformation floods on various social media, and continues to evolve in the era of AI-generated content (AIGC). The emerged misinformation with low creation cost and high deception poses significant threats to society. While recent studies leverage general-purpose multimodal large language models (MLLMs) to achieve remarkable results in detection, they encounter two critical limitations: (1) Insufficient reasoning, where general-purpose MLLMs often follow the uniform reasoning paradigm but generate inaccurate explanations and judgments, due to the lack of the task-specific knowledge of multimodal misinformation detection. (2) Reasoning biases, where a single thinking mode make detectors a suboptimal path for judgment, struggling to keep pace with the fast-growing and intricate multimodal misinformation. In this paper, we propose MMD-Thinker, a two-stage framework for multimodal misinformation detection through adaptive multi-dimensional thinking. First, we develop tailor-designed thinking mode for multimodal misinformation detection. Second, we adopt task-specific instruction tuning to inject the tailored thinking mode into general-purpose MLLMs. Third, we further leverage reinforcement learning strategy with a mixed advantage function, which incentivizes the reasoning capabilities in trajectories. Furthermore, we construct the multimodal misinformation reasoning (MMR) dataset, encompasses more than 8K image-text pairs with both reasoning processes and classification labels, to make progress in the relam of multimodal misinformation detection. Experimental results demonstrate that our proposed MMD-Thinker achieves state-of-the-art performance on both in-domain and out-of-domain benchmark datasets, while maintaining flexible inference and token usage. Code will be publicly available at Github.
Authors: Yu Wen, Shuyong Gao, Shuping Zhang, Miao Huang, Lili Tao, Han Yang, Haozhe Xing, Lihe Zhang, Boxue Hou
Abstract: Referring camouflaged object detection (Ref-COD) aims to identify hidden objects by incorporating reference information such as images and text descriptions. Previous research has transformed reference images with salient objects into one-dimensional prompts, yielding significant results. We explore ways to enhance performance through multi-context fusion of rich salient image features and camouflaged object features. Therefore, we propose RFMNet, which utilizes features from multiple encoding stages of the reference salient images and performs interactive fusion with the camouflage features at the corresponding encoding stages. Given that the features in salient object images contain abundant object-related detail information, performing feature fusion within local areas is more beneficial for detecting camouflaged objects. Therefore, we propose an Overlapped Windows Cross-attention mechanism to enable the model to focus more attention on the local information matching based on reference features. Besides, we propose the Referring Feature Aggregation (RFA) module to decode and segment the camouflaged objects progressively. Extensive experiments on the Ref-COD benchmark demonstrate that our method achieves state-of-the-art performance.
Authors: Yushuo Zheng, Jiangyong Ying, Huiyu Duan, Chunyi Li, Zicheng Zhang, Jing Liu, Xiaohong Liu, Guangtao Zhai
Abstract: Large multimodal models (LMMs) have demonstrated remarkable capabilities across a wide range of tasks, however their knowledge and abilities in the cross-view geo-localization and pose estimation domains remain unexplored, despite potential benefits for navigation, autonomous driving, outdoor robotics, \textit{etc}. To bridge this gap, we introduce \textbf{GeoX-Bench}, a comprehensive \underline{Bench}mark designed to explore and evaluate the capabilities of LMMs in \underline{cross}-view \underline{Geo}-localization and pose estimation. Specifically, GeoX-Bench contains 10,859 panoramic-satellite image pairs spanning 128 cities in 49 countries, along with corresponding 755,976 question-answering (QA) pairs. Among these, 42,900 QA pairs are designated for benchmarking, while the remaining are intended to enhance the capabilities of LMMs. Based on GeoX-Bench, we evaluate the capabilities of 25 state-of-the-art LMMs on cross-view geo-localization and pose estimation tasks, and further explore the empowered capabilities of instruction-tuning. Our benchmark demonstrate that while current LMMs achieve impressive performance in geo-localization tasks, their effectiveness declines significantly on the more complex pose estimation tasks, highlighting a critical area for future improvement, and instruction-tuning LMMs on the training data of GeoX-Bench can significantly improve the cross-view geo-sense abilities. The GeoX-Bench is available at \textcolor{magenta}{https://github.com/IntMeGroup/GeoX-Bench}.
Authors: Junlong Li, Huaiyuan Xu, Sijie Cheng, Kejun Wu, Kim-Hui Yap, Lap-Pui Chau, Yi Wang
Abstract: Driven by recent advances in vision language models (VLMs) and egocentric perception research, we introduce the concept of an egocentric procedural AI assistant (EgoProceAssist) tailored to step-by-step support daily procedural tasks in a first-person view. In this work, we start by identifying three core tasks: egocentric procedural error detection, egocentric procedural learning, and egocentric procedural question answering. These tasks define the essential functions of EgoProceAssist within a new taxonomy. Specifically, our work encompasses a comprehensive review of current techniques, relevant datasets, and evaluation metrics across these three core areas. To clarify the gap between the proposed EgoProceAssist and existing VLM-based AI assistants, we introduce novel experiments and provide a comprehensive evaluation of representative VLM-based methods. Based on these findings and our technical analysis, we discuss the challenges ahead and suggest future research directions. Furthermore, an exhaustive list of this study is publicly available in an active repository that continuously collects the latest work: https://github.com/z1oong/Building-Egocentric-Procedural-AI-Assistant
URLs: https://github.com/z1oong/Building-Egocentric-Procedural-AI-Assistant
Authors: Keshav Gupta, Akshat Sanghvi, Shreyas Reddy Palley, Astitva Srivastava, Charu Sharma, Avinash Sharma
Abstract: 3D Gaussian Splatting has emerged as a transformative technique in novel view synthesis, primarily due to its high rendering speed and photorealistic fidelity. However, its memory footprint scales rapidly with scene complexity, often reaching several gigabytes. Existing methods address this issue by introducing compression strategies that exploit primitive-level redundancy through similarity detection and quantization. We aim to surpass the compression limits of such methods by incorporating symmetry-aware techniques, specifically targeting mirror symmetries to eliminate redundant primitives. We propose a novel compression framework, \textbf{\textit{SymGS}}, introducing learnable mirrors into the scene, thereby eliminating local and global reflective redundancies for compression. Our framework functions as a plug-and-play enhancement to state-of-the-art compression methods, (e.g. HAC) to achieve further compression. Compared to HAC, we achieve $1.66 \times$ compression across benchmark datasets (upto $3\times$ on large-scale scenes). On an average, SymGS enables $\bf{108\times}$ compression of a 3DGS scene, while preserving rendering quality. The project page and supplementary can be found at \textbf{\color{cyan}{symgs.github.io}}
Authors: Lingfeng Zhang, Yuchen Zhang, Hongsheng Li, Haoxiang Fu, Yingbo Tang, Hangjun Ye, Long Chen, Xiaojun Liang, Xiaoshuai Hao, Wenbo Ding
Abstract: Vision-Language Models (VLMs), leveraging their powerful visual perception and reasoning capabilities, have been widely applied in Unmanned Aerial Vehicle (UAV) tasks. However, the spatial intelligence capabilities of existing VLMs in UAV scenarios remain largely unexplored, raising concerns about their effectiveness in navigating and interpreting dynamic environments. To bridge this gap, we introduce SpatialSky-Bench, a comprehensive benchmark specifically designed to evaluate the spatial intelligence capabilities of VLMs in UAV navigation. Our benchmark comprises two categories-Environmental Perception and Scene Understanding-divided into 13 subcategories, including bounding boxes, color, distance, height, and landing safety analysis, among others. Extensive evaluations of various mainstream open-source and closed-source VLMs reveal unsatisfactory performance in complex UAV navigation scenarios, highlighting significant gaps in their spatial capabilities. To address this challenge, we developed the SpatialSky-Dataset, a comprehensive dataset containing 1M samples with diverse annotations across various scenarios. Leveraging this dataset, we introduce Sky-VLM, a specialized VLM designed for UAV spatial reasoning across multiple granularities and contexts. Extensive experimental results demonstrate that Sky-VLM achieves state-of-the-art performance across all benchmark tasks, paving the way for the development of VLMs suitable for UAV scenarios. The source code is available at https://github.com/linglingxiansen/SpatialSKy.
Authors: Noam Tsfaty, Avishai Weizman, Liav Cohen, Moshe Tshuva, Yehudit Aperstein
Abstract: We address the challenge of detecting rare and diverse anomalies in surveillance videos using only video-level supervision. Our dual-backbone framework combines convolutional and transformer representations through top-k pooling, achieving 90.7% area under the curve (AUC) on the UCF-Crime dataset.
Authors: Zihan Li, Tengfei Wang, Wentian Gan, Hao Zhan, Xin Wang, Zongqian Zhan
Abstract: Lightweight building surface models are crucial for digital city, navigation, and fast geospatial analytics, yet conventional multi-view geometry pipelines remain cumbersome and quality-sensitive due to their reliance on dense reconstruction, meshing, and subsequent simplification. This work presents SF-Recon, a method that directly reconstructs lightweight building surfaces from multi-view images without post-hoc mesh simplification. We first train an initial 3D Gaussian Splatting (3DGS) field to obtain a view-consistent representation. Building structure is then distilled by a normal-gradient-guided Gaussian optimization that selects primitives aligned with roof and wall boundaries, followed by multi-view edge-consistency pruning to enhance structural sharpness and suppress non-structural artifacts without external supervision. Finally, a multi-view depth-constrained Delaunay triangulation converts the structured Gaussian field into a lightweight, structurally faithful building mesh. Based on a proposed SF dataset, the experimental results demonstrate that our SF-Recon can directly reconstruct lightweight building models from multi-view imagery, achieving substantially fewer faces and vertices while maintaining computational efficiency. Website:https://lzh282140127-cell.github.io/SF-Recon-project/
Authors: Kaiwen Wang, Kaili Zheng, Yiming Shi, Chenyi Guo, Ji Wu
Abstract: Multi-person human mesh recovery from a single image is a challenging task, hindered by the scarcity of in-the-wild training data. Prevailing in-the-wild human mesh pseudo-ground-truth (pGT) generation pipelines are single-person-centric, where each human is processed individually without joint optimization. This oversight leads to a lack of scene-level consistency, producing individuals with conflicting depths and scales within the same image. To address this, we introduce Depth-conditioned Translation Optimization (DTO), a novel optimization-based method that jointly refines the camera-space translations of all individuals in a crowd. By leveraging anthropometric priors on human height and depth cues from a monocular depth estimator, DTO solves for a scene-consistent placement of all subjects within a principled Maximum a posteriori (MAP) framework. Applying DTO to the 4D-Humans dataset, we construct DTO-Humans, a new large-scale pGT dataset of 0.56M high-quality, scene-consistent multi-person images, featuring dense crowds with an average of 4.8 persons per image. Furthermore, we propose Metric-Aware HMR, an end-to-end network that directly estimates human mesh and camera parameters in metric scale. This is enabled by a camera branch and a novel relative metric loss that enforces plausible relative scales. Extensive experiments demonstrate that our method achieves state-of-the-art performance on relative depth reasoning and human mesh recovery. Code and data will be released publicly.
Authors: Jongha Kim, Minseong Bae, Sanghyeok Lee, Jinsung Yoon, Hyunwoo J. Kim
Abstract: Table images present unique challenges for effective and efficient understanding due to the need for question-specific focus and the presence of redundant background regions. Existing Multimodal Large Language Model (MLLM) approaches often overlook these characteristics, resulting in uninformative and redundant visual representations. To address these issues, we aim to generate visual features that are both informative and compact to improve table understanding. We first propose progressive question conditioning, which injects the question into Vision Transformer layers with gradually increasing frequency, considering each layer's capacity to handle additional information, to generate question-aware visual features. To reduce redundancy, we introduce a pruning strategy that discards background tokens, thereby improving efficiency. To mitigate information loss from pruning, we further propose token focusing, a training strategy that encourages the model to concentrate essential information in the retained tokens. By combining these approaches, we present TabFlash, an efficient and effective MLLM for table understanding. TabFlash achieves state-of-the-art performance, outperforming both open-source and proprietary MLLMs, while requiring 27% less FLOPs and 30% less memory usage compared to the second-best MLLM.
Authors: Yunjie Yu, Jingchen Wu, Junchen Zhu, Chunze Lin, Guibin Chen
Abstract: Artistic design such as poster design often demands rapid yet precise modification of textual content while preserving visual harmony and typographic intent, especially across diverse font styles. Although modern image editing models have grown increasingly powerful, they still fall short in fine-grained, font-aware text manipulation, limiting their utility in professional design workflows such as poster editing. To address this issue, we present SkyReels-Text, a novel font-controllable framework for precise poster text editing. Our method enables simultaneous editing of multiple text regions, each rendered in distinct typographic styles, while preserving the visual appearance of non-edited regions. Notably, our model requires neither font labels nor fine-tuning during inference: users can simply provide cropped glyph patches corresponding to their desired typography, even if the font is not included in any standard library. Extensive experiments on multiple datasets, including handwrittent text benchmarks, SkyReels-Text achieves state-of-the-art performance in both text fidelity and visual realism, offering unprecedented control over font families, and stylistic nuances. This work bridges the gap between general-purpose image editing and professional-grade typographic design.
Authors: Enhui Ma, Lijun Zhou, Tao Tang, Jiahuan Zhang, Junpeng Jiang, Zhan Zhang, Dong Han, Kun Zhan, Xueyang Zhang, XianPeng Lang, Haiyang Sun, Xia Zhou, Di Lin, Kaicheng Yu
Abstract: End-to-end planning methods are the de facto standard of the current autonomous driving system, while the robustness of the data-driven approaches suffers due to the notorious long-tail problem (i.e., rare but safety-critical failure cases). In this work, we explore whether recent diffusion-based video generation methods (a.k.a. world models), paired with structured 3D layouts, can enable a fully automated pipeline to self-correct such failure cases. We first introduce an agent to simulate the role of product manager, dubbed PM-Agent, which formulates data requirements to collect data similar to the failure cases. Then, we use a generative model that can simulate both data collection and annotation. However, existing generative models struggle to generate high-fidelity data conditioned on 3D layouts. To address this, we propose DriveSora, which can generate spatiotemporally consistent videos aligned with the 3D annotations requested by PM-Agent. We integrate these components into our self-correcting agentic system, CorrectAD. Importantly, our pipeline is an end-to-end model-agnostic and can be applied to improve any end-to-end planner. Evaluated on both nuScenes and a more challenging in-house dataset across multiple end-to-end planners, CorrectAD corrects 62.5% and 49.8% of failure cases, reducing collision rates by 39% and 27%, respectively.
Authors: Kaiwen Cai, Xinze Liu, Xia Zhou, Hengtong Hu, Jie Xiang, Luyao Zhang, Xueyang Zhang, Kun Zhan, Yifei Zhan, Xianpeng Lang
Abstract: The generation of realistic LiDAR point clouds plays a crucial role in the development and evaluation of autonomous driving systems. Although recent methods for 3D LiDAR point cloud generation have shown significant improvements, they still face notable limitations, including the lack of sequential generation capabilities and the inability to produce accurately positioned foreground objects and realistic backgrounds. These shortcomings hinder their practical applicability. In this paper, we introduce DriveLiDAR4D, a novel LiDAR generation pipeline consisting of multimodal conditions and a novel sequential noise prediction model LiDAR4DNet, capable of producing temporally consistent LiDAR scenes with highly controllable foreground objects and realistic backgrounds. To the best of our knowledge, this is the first work to address the sequential generation of LiDAR scenes with full scene manipulation capability in an end-to-end manner. We evaluated DriveLiDAR4D on the nuScenes and KITTI datasets, where we achieved an FRD score of 743.13 and an FVD score of 16.96 on the nuScenes dataset, surpassing the current state-of-the-art (SOTA) method, UniScene, with an performance boost of 37.2% in FRD and 24.1% in FVD, respectively.
Authors: Narthana Sivalingam, Santhirarajah Sivasthigan, Thamayanthi Mahendranathan, G. M. R. I. Godaliyadda, M. P. B. Ekanayake, H. M. V. R. Herath
Abstract: Group activity detection in multi-person scenes is challenging due to complex human interactions, occlusions, and variations in appearance over time. This work presents a computer vision based framework for group activity recognition and action spotting using a combination of deep learning models and graph based relational reasoning. The system first applies Mask R-CNN to obtain accurate actor localization through bounding boxes and instance masks. Multiple backbone networks, including Inception V3, MobileNet, and VGG16, are used to extract feature maps, and RoIAlign is applied to preserve spatial alignment when generating actor specific features. The mask information is then fused with the feature maps to obtain refined masked feature representations for each actor. To model interactions between individuals, we construct Actor Relation Graphs that encode appearance similarity and positional relations using methods such as normalized cross correlation, sum of absolute differences, and dot product. Graph Convolutional Networks operate on these graphs to reason about relationships and predict both individual actions and group level activities. Experiments on the Collective Activity dataset demonstrate that the combination of mask based feature refinement, robust similarity search, and graph neural network reasoning leads to improved recognition performance across both crowded and non crowded scenarios. This approach highlights the potential of integrating segmentation, feature extraction, and relational graph reasoning for complex video understanding tasks.
Authors: Ori Meiraz, Sharon Shalev, Avishai Weizman
Abstract: This paper presents a novel Mixture-of-Experts framework for object detection, incorporating adaptive routing among multiple YOLOv9-T experts to enable dynamic feature specialization and achieve higher mean Average Precision (mAP) and Average Recall (AR) compared to a single YOLOv9-T model.
Authors: Lucas Gabriel Telesco, Danila Nejamkin, Estefan\'ia Mata, Francisco Filizzola, Kevin Wignall, Luc\'ia Franco Troilo, Mar\'ia de los Angeles Cenoz, Melissa Thompson, Mercedes Legu\'ia, Ignacio Larrabide, Jos\'e Ignacio Orlando
Abstract: Retinal image quality assessment (RIQA) supports computer-aided diagnosis of eye diseases. However, most tools classify only overall image quality, without indicating acquisition defects to guide recapture. This gap is mainly due to the high cost of detailed annotations. In this paper, we aim to mitigate this limitation by introducing a hybrid semi-supervised learning approach that combines manual labels for overall quality with pseudo-labels of quality details within a multi-task framework. Our objective is to obtain more interpretable RIQA models without requiring extensive manual labeling. Pseudo-labels are generated by a Teacher model trained on a small dataset and then used to fine-tune a pre-trained model in a multi-task setting. Using a ResNet-18 backbone, we show that these weak annotations improve quality assessment over single-task baselines (F1: 0.875 vs. 0.863 on EyeQ, and 0.778 vs. 0.763 on DeepDRiD), matching or surpassing existing methods. The multi-task model achieved performance statistically comparable to the Teacher for most detail prediction tasks (p > 0.05). In a newly annotated EyeQ subset released with this paper, our model performed similarly to experts, suggesting that pseudo-label noise aligns with expert variability. Our main finding is that the proposed semi-supervised approach not only improves overall quality assessment but also provides interpretable feedback on capture conditions (illumination, clarity, contrast). This enhances interpretability at no extra manual labeling cost and offers clinically actionable outputs to guide image recapture.
Authors: Fei Kong
Abstract: Recently, the Denoising Diffusion Codebook Models (DDCM) was proposed. DDCM leverages the Denoising Diffusion Probabilistic Model (DDPM) and replaces the random noise in the backward process with noise sampled from specific sets according to a predefined rule, thereby enabling image compression. However, DDCM cannot be applied to methods other than DDPM. In this paper, we propose the generalized Denoising Diffusion Compression Model (gDDCM), which extends DDCM to mainstream diffusion models and their variants, including DDPM, Score-Based Models, Consistency Models, and Rectified Flow. We evaluate our method on CIFAR-10 and LSUN Bedroom datasets. Experimental results demonstrate that our approach successfully generalizes DDCM to the aforementioned models and achieves improved performance.
Authors: Nikos Theodoridis, Tim Brophy, Reenu Mohandas, Ganesh Sistu, Fiachra Collins, Anthony Scanlan, Ciaran Eising
Abstract: The remarkable progress of Vision-Language Models (VLMs) on a variety of tasks has raised interest in their application to automated driving. However, for these models to be trusted in such a safety-critical domain, they must first possess robust perception capabilities, i.e., they must be capable of understanding a traffic scene, which can often be highly complex, with many things happening simultaneously. Moreover, since critical objects and agents in traffic scenes are often at long distances, we require systems with not only strong perception capabilities at close distances (up to 20 meters), but also at long (30+ meters) range. Therefore, it is important to evaluate the perception capabilities of these models in isolation from other skills like reasoning or advanced world knowledge. Distance-Annotated Traffic Perception Question Answering (DTPQA) is a Visual Question Answering (VQA) benchmark designed specifically for this purpose: it can be used to evaluate the perception systems of VLMs in traffic scenarios using trivial yet crucial questions relevant to driving decisions. It consists of two parts: a synthetic benchmark (DTP-Synthetic) created using a simulator, and a real-world benchmark (DTP-Real) built on top of existing images of real traffic scenes. Additionally, DTPQA includes distance annotations, i.e., how far the object in question is from the camera. More specifically, each DTPQA sample consists of (at least): (a) an image, (b) a question, (c) the ground truth answer, and (d) the distance of the object in question, enabling analysis of how VLM performance degrades with increasing object distance. In this article, we provide the dataset itself along with the Python scripts used to create it, which can be used to generate additional data of the same kind.
Authors: Yuchen Bao, Yiting Wang, Wenjian Huang, Haowei Wang, Shen Chen, Taiping Yao, Shouhong Ding, Jianguo Zhang
Abstract: Scene Text Editing (STE) aims to naturally modify text in images while preserving visual consistency, the decisive factors of which can be divided into three parts, i.e., text style, text content, and background. Previous methods have struggled with incomplete disentanglement of editable attributes, typically addressing only one aspect - such as editing text content - thus limiting controllability and visual consistency. To overcome these limitations, we propose TripleFDS, a novel framework for STE with disentangled modular attributes, and an accompanying dataset called SCB Synthesis. SCB Synthesis provides robust training data for triple feature disentanglement by utilizing the "SCB Group", a novel construct that combines three attributes per image to generate diverse, disentangled training groups. Leveraging this construct as a basic training unit, TripleFDS first disentangles triple features, ensuring semantic accuracy through inter-group contrastive regularization and reducing redundancy through intra-sample multi-feature orthogonality. In the synthesis phase, TripleFDS performs feature remapping to prevent "shortcut" phenomena during reconstruction and mitigate potential feature leakage. Trained on 125,000 SCB Groups, TripleFDS achieves state-of-the-art image fidelity (SSIM of 44.54) and text accuracy (ACC of 93.58%) on the mainstream STE benchmarks. Besides superior performance, the more flexible editing of TripleFDS supports new operations such as style replacement and background transfer. Code: https://github.com/yusenbao01/TripleFDS
Authors: Jinkun Zhao, Lei Huang, Wenjun Wu
Abstract: With the rapid advancement of Large Models, numerous text-and-vision-fused Multimodal Large Models (MLMs) have emerged. However, these MLMs remain susceptible to informational interference in visual perception, particularly in color perception, which introduces an additional risk of hallucination. To validate this hypothesis, we introduce the "What Color Is It" dataset, a novel benchmark constructed using a simple method to trigger single-modality visual hallucination in MLMs. Based on this dataset, we further investigate the underlying causes of hallucination in the visual modality of MLMs and propose potential solutions to enhance their robustness.
Authors: Mykola Lavreniuk, Nataliia Kussul, Andrii Shelestov, Yevhenii Salii, Volodymyr Kuzin, Sergii Skakun, Zoltan Szantoi
Abstract: Accurate delineation of agricultural field boundaries from satellite imagery is essential for land management and crop monitoring, yet existing methods often produce incomplete boundaries, merge adjacent fields, and struggle to scale. We present the Delineate Anything Flow (DelAnyFlow) methodology, a resolution-agnostic approach for large-scale field boundary mapping. DelAnyFlow combines the DelAny instance segmentation model, based on a YOLOv11 backbone and trained on the large-scale Field Boundary Instance Segmentation-22M (FBIS 22M) dataset, with a structured post-processing, merging, and vectorization sequence to generate topologically consistent vector boundaries. FBIS 22M, the largest dataset of its kind, contains 672,909 multi-resolution image patches (0.25-10m) and 22.9million validated field instances. The DelAny model delivers state-of-the-art accuracy with over 100% higher mAP and 400x faster inference than SAM2. DelAny demonstrates strong zero-shot generalization and supports national-scale applications: using Sentinel 2 data for 2024, DelAnyFlow generated a complete field boundary layer for Ukraine (603,000km2) in under six hours on a single workstation. DelAnyFlow outputs significantly improve boundary completeness relative to operational products from Sinergise Solutions and NASA Harvest, particularly in smallholder and fragmented systems (0.25-1ha). For Ukraine, DelAnyFlow delineated 3.75M fields at 5m and 5.15M at 2.5m, compared to 2.66M detected by Sinergise Solutions and 1.69M by NASA Harvest. This work delivers a scalable, cost-effective methodology for field delineation in regions lacking digital cadastral data. A project landing page with links to model weights, code, national-scale vector outputs, and dataset is available at https://lavreniuk.github.io/Delineate-Anything/.
Authors: Xingming Long, Jie Zhang, Shiguang Shan, Xilin Chen
Abstract: Most research on hallucinations in Large Vision-Language Models (LVLMs) focuses on factual description tasks that prohibit any output absent from the image. However, little attention has been paid to hallucinations in voluntary imagination tasks, e.g., story writing, where the models are expected to generate novel content beyond the given image. In these tasks, it is inappropriate to simply regard such imagined novel content as hallucinations. To address this limitation, we introduce Voluntary-imagined Object Presence Evaluation (VOPE)-a novel method to assess LVLMs' hallucinations in voluntary imagination tasks via presence evaluation. Specifically, VOPE poses recheck-based questions to evaluate how an LVLM interprets the presence of the imagined objects in its own response. The consistency between the model's interpretation and the object's presence in the image is then used to determine whether the model hallucinates when generating the response. We apply VOPE to several mainstream LVLMs and hallucination mitigation methods, revealing two key findings: (1) most LVLMs hallucinate heavily during voluntary imagination, and their performance in presence evaluation is notably poor on imagined objects; (2) existing hallucination mitigation methods show limited effect in voluntary imagination tasks, making this an important direction for future research.
Authors: Lorenzo Olearo, Giulio Vigan\`o, Daniele Baieri, Filippo Maggioli, Simone Melzi
Abstract: We introduce a novel neural representation for maps between 3D shapes based on flow-matching models, which is computationally efficient and supports cross-representation shape matching without large-scale training or data-driven procedures. 3D shapes are represented as the probability distribution induced by a continuous and invertible flow mapping from a fixed anchor distribution. Given a source and a target shape, the composition of the inverse flow (source to anchor) with the forward flow (anchor to target), we continuously map points between the two surfaces. By encoding the shapes with a pointwise task-tailored embedding, this construction provides an invertible and modality-agnostic representation of maps between shapes across point clouds, meshes, signed distance fields (SDFs), and volumetric data. The resulting representation consistently achieves high coverage and accuracy across diverse benchmarks and challenging settings in shape matching. Beyond shape matching, our framework shows promising results in other tasks, including UV mapping and registration of raw point cloud scans of human bodies.
Authors: Rui Zuo, Qinyue Tong, Zhe-Ming Lu, Ziqian Lu
Abstract: With the rapid advancement of artificial intelligence-generated content (AIGC) technologies, including multimodal large language models (MLLMs) and diffusion models, image generation and manipulation have become remarkably effortless. Existing image forgery detection and localization (IFDL) methods often struggle to generalize across diverse datasets and offer limited interpretability. Nowadays, MLLMs demonstrate strong generalization potential across diverse vision-language tasks, and some studies introduce this capability to IFDL via large-scale training. However, such approaches cost considerable computational resources, while failing to reveal the inherent generalization potential of vanilla MLLMs to address this problem. Inspired by this observation, we propose Foresee, a training-free MLLM-based pipeline tailored for image forgery analysis. It eliminates the need for additional training and enables a lightweight inference process, while surpassing existing MLLM-based methods in both tamper localization accuracy and the richness of textual explanations. Foresee employs a type-prior-driven strategy and utilizes a Flexible Feature Detector (FFD) module to specifically handle copy-move manipulations, thereby effectively unleashing the potential of vanilla MLLMs in the forensic domain. Extensive experiments demonstrate that our approach simultaneously achieves superior localization accuracy and provides more comprehensive textual explanations. Moreover, Foresee exhibits stronger generalization capability, outperforming existing IFDL methods across various tampering types, including copy-move, splicing, removal, local enhancement, deepfake, and AIGC-based editing. The code will be released in the final version.
Authors: Adam Hazimeh, Ke Wang, Mark Collier, Gilles Baechler, Efi Kokiopoulou, Pascal Frossard
Abstract: Multimedia documents such as slide presentations and posters are designed to be interactive and easy to modify. Yet, they are often distributed in a static raster format, which limits editing and customization. Restoring their editability requires converting these raster images back into structured vector formats. However, existing geometric raster-vectorization methods, which rely on low-level primitives like curves and polygons, fall short at this task. Specifically, when applied to complex documents like slides, they fail to preserve the high-level structure, resulting in a flat collection of shapes where the semantic distinction between image and text elements is lost. To overcome this limitation, we address the problem of semantic document derendering by introducing SliDer, a novel framework that uses Vision-Language Models (VLMs) to derender slide images as compact and editable Scalable Vector Graphic (SVG) representations. SliDer detects and extracts attributes from individual image and text elements in a raster input and organizes them into a coherent SVG format. Crucially, the model iteratively refines its predictions during inference in a process analogous to human design, generating SVG code that more faithfully reconstructs the original raster upon rendering. Furthermore, we introduce Slide2SVG, a novel dataset comprising raster-SVG pairs of slide documents curated from real-world scientific presentations, to facilitate future research in this domain. Our results demonstrate that SliDer achieves a reconstruction LPIPS of 0.069 and is favored by human evaluators in 82.9% of cases compared to the strongest zero-shot VLM baseline.
Authors: Lipeng Wang, Hongxing Fan, Haohua Chen, Zehuan Huang, Lu Sheng
Abstract: Generating high-quality human interactions holds significant value for applications like virtual reality and robotics. However, existing methods often fail to preserve unique individual characteristics or fully adhere to textual descriptions. To address these challenges, we introduce InterMoE, a novel framework built on a Dynamic Temporal-Selective Mixture of Experts. The core of InterMoE is a routing mechanism that synergistically uses both high-level text semantics and low-level motion context to dispatch temporal motion features to specialized experts. This allows experts to dynamically determine the selection capacity and focus on critical temporal features, thereby preserving specific individual characteristic identities while ensuring high semantic fidelity. Extensive experiments show that InterMoE achieves state-of-the-art performance in individual-specific high-fidelity 3D human interaction generation, reducing FID scores by 9% on the InterHuman dataset and 22% on InterX.
Authors: Jae Joong Lee
Abstract: Recent vision-language models (VLMs) such as CLIP, OpenCLIP, EVA02-CLIP and SigLIP achieve strong zero-shot performance, but it is unclear how reliably they respond to controlled linguistic perturbations. We introduce Language-Guided Invariance Probing (LGIP), a benchmark that measures (i) invariance to meaning-preserving paraphrases and (ii) sensitivity to meaning-changing semantic flips in image-text matching. Using 40k MS COCO images with five human captions each, we automatically generate paraphrases and rule-based flips that alter object category, color or count, and summarize model behavior with an invariance error, a semantic sensitivity gap and a positive-rate statistic. Across nine VLMs, EVA02-CLIP and large OpenCLIP variants lie on a favorable invariance-sensitivity frontier, combining low paraphrase-induced variance with consistently higher scores for original captions than for their flipped counterparts. In contrast, SigLIP and SigLIP2 show much larger invariance error and often prefer flipped captions to the human descriptions, especially for object and color edits. These failures are largely invisible to standard retrieval metrics, indicating that LGIP provides a model-agnostic diagnostic for the linguistic robustness of VLMs beyond conventional accuracy scores.
Authors: Wenyu Zhang, Yao Tong, Yiqiu Liu, Rui Cao
Abstract: Urban villages (UVs), informal settlements embedded within China's urban fabric, have undergone widespread demolition and redevelopment in recent decades. However, there remains a lack of systematic evaluation of whether the demolished land has been effectively reused, raising concerns about the efficacy and sustainability of current redevelopment practices. To address the gap, this study proposes a deep learning-based framework to monitor the spatiotemporal changes of UVs in China. Specifically, semantic segmentation of multi-temporal remote sensing imagery is first used to map evolving UV boundaries, and then post-demolition land use is classified into six categories based on the "remained-demolished-redeveloped" phase: incomplete demolition, vacant land, construction sites, buildings, green spaces, and others. Four representative cities from China's four economic regions were selected as the study areas, i.e., Guangzhou (East), Zhengzhou (Central), Xi'an (West), and Harbin (Northeast). The results indicate: 1) UV redevelopment processes were frequently prolonged; 2) redevelopment transitions primarily occurred in peripheral areas, whereas urban cores remained relatively stable; and 3) three spatiotemporal transformation pathways, i.e., synchronized redevelopment, delayed redevelopment, and gradual optimization, were revealed. This study highlights the fragmented, complex and nonlinear nature of UV redevelopment, underscoring the need for tiered and context-sensitive planning strategies. By linking spatial dynamics with the context of redevelopment policies, the findings offer valuable empirical insights that support more inclusive, efficient, and sustainable urban renewal, while also contributing to a broader global understanding of informal settlement transformations.
Authors: Jeffrey Wen, Rizwan Ahmad, Philip Schniter
Abstract: In ill-posed imaging inverse problems, uncertainty quantification remains a fundamental challenge, especially in safety-critical applications. Recently, conformal prediction has been used to quantify the uncertainty that the inverse problem contributes to downstream tasks like image classification, image quality assessment, fat mass quantification, etc. While existing works handle only a scalar estimation target, practical applications often involve multiple targets. In response, we propose an asymptotically minimax approach to multi-target conformal prediction that provides tight prediction intervals while ensuring joint marginal coverage. We then outline how our minimax approach can be applied to multi-metric blind image quality assessment, multi-task uncertainty quantification, and multi-round measurement acquisition. Finally, we numerically demonstrate the benefits of our minimax method, relative to existing multi-target conformal prediction methods, using both synthetic and magnetic resonance imaging (MRI) data.
Authors: Farhin Farhad Riya, Shahinul Hoque, Jinyuan Stella Sun, Olivera Kotevska
Abstract: As machine learning models are increasingly deployed in safety-critical domains, visual explanation techniques have become essential tools for supporting transparency. In this work, we reveal a new class of attacks that compromise model interpretability without affecting accuracy. Specifically, we show that small color perturbations applied by adversarial clients in a federated learning setting can shift a model's saliency maps away from semantically meaningful regions while keeping the prediction unchanged. The proposed saliency-aware attack framework, called Chromatic Perturbation Module, systematically crafts adversarial examples by altering the color contrast between foreground and background in a way that disrupts explanation fidelity. These perturbations accumulate across training rounds, poisoning the global model's internal feature attributions in a stealthy and persistent manner. Our findings challenge a common assumption in model auditing that correct predictions imply faithful explanations and demonstrate that interpretability itself can be an attack surface. We evaluate this vulnerability across multiple datasets and show that standard training pipelines are insufficient to detect or mitigate explanation degradation, especially in the federated learning setting, where subtle color perturbations are harder to discern. Our attack reduces peak activation overlap in Grad-CAM explanations by up to 35% while preserving classification accuracy above 96% on all evaluated datasets.
Authors: Yuanchao Wang, Tian Qin, Eduardo Valle, Bruno Abrahao
Abstract: Out-of-distribution (OOD) detection is critical for deploying image classifiers in safety-sensitive environments, yet existing detectors often struggle when OOD samples are semantically similar to the in-distribution (ID) classes. We present BootOOD, a fully self-supervised OOD detection framework that bootstraps exclusively from ID data and is explicitly designed to handle semantically challenging OOD samples. BootOOD synthesizes pseudo-OOD features through simple transformations of ID representations and leverages Neural Collapse (NC), where ID features cluster tightly around class means with consistent feature norms. Unlike prior approaches that aim to constrain OOD features into subspaces orthogonal to the collapsed ID means, BootOOD introduces a lightweight auxiliary head that performs radius-based classification on feature norms. This design decouples OOD detection from the primary classifier and imposes a relaxed requirement: OOD samples are learned to have smaller feature norms than ID features, which is easier to satisfy when ID and OOD are semantically close. Experiments on CIFAR-10, CIFAR-100, and ImageNet-200 show that BootOOD outperforms prior post-hoc methods, surpasses training-based methods without outlier exposure, and is competitive with state-of-the-art outlier-exposure approaches while maintaining or improving ID accuracy.
Authors: Md. Iqbal Hossain, Afia Sajeeda, Neeresh Kumar Perla, Ming Shao
Abstract: The advent of multimodal deep learning models, such as CLIP, has unlocked new frontiers in a wide range of applications, from image-text understanding to classification tasks. However, these models are not safe for adversarial attacks, particularly backdoor attacks, which can subtly manipulate model behavior. Moreover, existing defense methods typically involve training from scratch or fine-tuning using a large dataset without pinpointing the specific labels that are affected. In this study, we introduce an innovative strategy to enhance the robustness of multimodal contrastive learning models against such attacks. In particular, given a poisoned CLIP model, our approach can identify the backdoor trigger and pinpoint the victim samples and labels in an efficient manner. To that end, an image segmentation ``oracle'' is introduced as the supervisor for the output of the poisoned CLIP. We develop two algorithms to rectify the poisoned model: (1) differentiating between CLIP and Oracle's knowledge to identify potential triggers; (2) pinpointing affected labels and victim samples, and curating a compact fine-tuning dataset. With this knowledge, we are allowed to rectify the poisoned CLIP model to negate backdoor effects. Extensive experiments on visual recognition benchmarks demonstrate our strategy is effective in CLIP-based backdoor defense.
Authors: Sining Chen, Xiao Xiang Zhu
Abstract: Monocular height estimation plays a critical role in 3D perception for remote sensing, offering a cost-effective alternative to multi-view or LiDAR-based methods. While deep learning has significantly advanced the capabilities of monocular height estimation, these methods remain fundamentally limited by the availability of labeled data, which are expensive and labor-intensive to obtain at scale. The scarcity of high-quality annotations hinders the generalization and performance of existing models. To overcome this limitation, we propose leveraging large volumes of unlabeled data through a semi-supervised learning framework, enabling the model to extract informative cues from unlabeled samples and improve its predictive performance. In this work, we introduce TSE-Net, a self-training pipeline for semi-supervised monocular height estimation. The pipeline integrates teacher, student, and exam networks. The student network is trained on unlabeled data using pseudo-labels generated by the teacher network, while the exam network functions as a temporal ensemble of the student network to stabilize performance. The teacher network is formulated as a joint regression and classification model: the regression branch predicts height values that serve as pseudo-labels, and the classification branch predicts height value classes along with class probabilities, which are used to filter pseudo-labels. Height value classes are defined using a hierarchical bi-cut strategy to address the inherent long-tailed distribution of heights, and the predicted class probabilities are calibrated with a Plackett-Luce model to reflect the expected accuracy of pseudo-labels. We evaluate the proposed pipeline on three datasets spanning different resolutions and imaging modalities. Codes are available at https://github.com/zhu-xlab/tse-net.
Authors: Ziyang Huang, Jiagang Chen, Jin Liu, Shunping Ji
Abstract: 3D Gaussian Splatting (3DGS) has emerged as a leading framework for novel view synthesis, yet its core optimization challenges remain underexplored. We identify two key issues in 3DGS optimization: entrapment in suboptimal local optima and insufficient convergence quality. To address these, we propose Opt3DGS, a robust framework that enhances 3DGS through a two-stage optimization process of adaptive exploration and curvature-guided exploitation. In the exploration phase, an Adaptive Weighted Stochastic Gradient Langevin Dynamics (SGLD) method enhances global search to escape local optima. In the exploitation phase, a Local Quasi-Newton Direction-guided Adam optimizer leverages curvature information for precise and efficient convergence. Extensive experiments on diverse benchmark datasets demonstrate that Opt3DGS achieves state-of-the-art rendering quality by refining the 3DGS optimization process without modifying its underlying representation.
Authors: Linhan Zhou, Shuang Li, Neng Dong, Yonghang Tai, Yafei Zhang, Huafeng Li
Abstract: Person re-identification (ReID) aims to retrieve target pedestrian images given either visual queries (image-to-image, I2I) or textual descriptions (text-to-image, T2I). Although both tasks share a common retrieval objective, they pose distinct challenges: I2I emphasizes discriminative identity learning, while T2I requires accurate cross-modal semantic alignment. Existing methods often treat these tasks separately, which may lead to representation entanglement and suboptimal performance. To address this, we propose a unified framework named Hierarchical Prompt Learning (HPL), which leverages task-aware prompt modeling to jointly optimize both tasks. Specifically, we first introduce a Task-Routed Transformer, which incorporates dual classification tokens into a shared visual encoder to route features for I2I and T2I branches respectively. On top of this, we develop a hierarchical prompt generation scheme that integrates identity-level learnable tokens with instance-level pseudo-text tokens. These pseudo-tokens are derived from image or text features via modality-specific inversion networks, injecting fine-grained, instance-specific semantics into the prompts. Furthermore, we propose a Cross-Modal Prompt Regularization strategy to enforce semantic alignment in the prompt token space, ensuring that pseudo-prompts preserve source-modality characteristics while enhancing cross-modal transferability. Extensive experiments on multiple ReID benchmarks validate the effectiveness of our method, achieving state-of-the-art performance on both I2I and T2I tasks.
Authors: Yinuo Xu, Yan Cui, Mingyao Li, Zhi Huang
Abstract: Identifying cell types and subtypes from routine histopathology images is essential for improving the computational understanding of human disease. Existing tile-based models can capture detailed nuclear morphology but often fail to incorporate the broader tissue context that influences a cell's function and identity. In addition, available human annotations are typically coarse-grained and unevenly distributed across studies, making fine-grained subtype-level supervision difficult to obtain. To address these limitations, we introduce NuClass, a pathologist workflow inspired framework for cell-wise multi-scale integration of nuclear morphology and microenvironmental context. NuClass includes two main components: Path local, which focuses on nuclear morphology from 224-by-224 pixel crops, and Path global, which models the surrounding 1024-by-1024 pixel neighborhood. A learnable gating module adaptively balances local detail and contextual cues. To encourage complementary learning, we incorporate an uncertainty-guided objective that directs the global path to prioritize regions where the local path is uncertain. We also provide calibrated confidence estimates and Grad-CAM visualizations to enhance interpretability. To overcome the lack of high-quality annotations, we construct a marker-guided dataset from Xenium spatial transcriptomics assays, yielding single-cell resolution labels for more than two million cells across eight organs and 16 classes. Evaluated on three fully held-out cohorts, NuClass achieves up to 96 percent F1 for its best-performing class, outperforming strong baselines. Our results show that multi-scale, uncertainty-aware fusion can bridge the gap between slide-level pathological foundation models and reliable, cell-level phenotype prediction.
Authors: Haotian Dong, Ye Li, Rongwei Lu, Chen Tang, Shu-Tao Xia, Zhi Wang
Abstract: Visual autoregressive (AR) generation models have demonstrated strong potential for image generation, yet their next-token-prediction paradigm introduces considerable inference latency. Although speculative decoding (SD) has been proven effective for accelerating visual AR models, its "draft one step, then verify one step" paradigm prevents a direct reduction of the forward passes, thus restricting acceleration potential. Motivated by the visual token interchangeability, we for the first time to explore verification skipping in the SD process of visual AR model generation to explicitly cut the number of target model forward passes, thereby reducing inference latency. Based on an analysis of the drafting stage's characteristics, we observe that verification redundancy and stale feature reusability are key factors to retain generation quality and speedup for verification-free steps. Inspired by these two observations, we propose a novel SD framework VVS to accelerate visual AR generation via partial verification skipping, which integrates three complementary modules: (1) a verification-free token selector with dynamical truncation, (2) token-level feature caching and reuse, and (3) fine-grained skipped step scheduling. Consequently, VVS reduces the number of target model forward passes by a factor of $2.8\times$ relative to vanilla AR decoding while maintaining competitive generation quality, offering a superior speed-quality trade-off over conventional SD frameworks and revealing strong potential to reshape the SD paradigm.
Authors: Xin Xu, Hao Liu, Wei Liu, Wei Wang, Jiayi Wu, Kui Jiang
Abstract: Low-Light Image Enhancement (LLIE) task aims at improving contrast while restoring details and textures for images captured in low-light conditions. HVI color space has made significant progress in this task by enabling precise decoupling of chrominance and luminance. However, for the interaction of chrominance and luminance branches, substantial distributional differences between the two branches prevalent in natural images limit complementary feature extraction, and luminance errors are propagated to chrominance channels through the nonlinear parameter. Furthermore, for interaction between different chrominance branches, images with large homogeneous-color regions usually exhibit weak correlation between chrominance branches due to concentrated distributions. Traditional pixel-wise losses exploit strong inter-branch correlations for co-optimization, causing gradient conflicts in weakly correlated regions. Therefore, we propose an Inter-Chrominance and Luminance Interaction (ICLR) framework including a Dual-stream Interaction Enhancement Module (DIEM) and a Covariance Correction Loss (CCL). The DIEM improves the extraction of complementary information from two dimensions, fusion and enhancement, respectively. The CCL utilizes luminance residual statistics to penalize chrominance errors and balances gradient conflicts by constraining chrominance branches covariance. Experimental results on multiple datasets show that the proposed ICLR framework outperforms state-of-the-art methods.
Authors: Marianne Rakic, Andrew Hoopes, S. Mazdak Abulnaga, Mert R. Sabuncu, John V. Guttag, Adrian V. Dalca
Abstract: Deformable templates, or atlases, are images that represent a prototypical anatomy for a population, and are often enhanced with probabilistic anatomical label maps. They are commonly used in medical image analysis for population studies and computational anatomy tasks such as registration and segmentation. Because developing a template is a computationally expensive process, relatively few templates are available. As a result, analysis is often conducted with sub-optimal templates that are not truly representative of the study population, especially when there are large variations within this population. We propose a machine learning framework that uses convolutional registration neural networks to efficiently learn a function that outputs templates conditioned on subject-specific attributes, such as age and sex. We also leverage segmentations, when available, to produce anatomical segmentation maps for the resulting templates. The learned network can also be used to register subject images to the templates. We demonstrate our method on a compilation of 3D brain MRI datasets, and show that it can learn high-quality templates that are representative of populations. We find that annotated conditional templates enable better registration than their unlabeled unconditional counterparts, and outperform other templates construction methods.
Authors: Kesi Xu, Eleni Chiou, Ali Varamesh, Laura Acqualagna, Nasir Rajpoot
Abstract: Accurate nuclei detection and classification are fundamental to computational pathology, yet existing approaches are hindered by reliance on detailed expert annotations and insufficient use of tissue context. We present Tissue-Aware Nuclei Detection (TAND), a novel framework achieving joint nuclei detection and classification using point-level supervision enhanced by tissue mask conditioning. TAND couples a ConvNeXt-based encoder-decoder with a frozen Virchow-2 tissue segmentation branch, where semantic tissue probabilities selectively modulate the classification stream through a novel multi-scale Spatial Feature-wise Linear Modulation (Spatial-FiLM). On the PUMA benchmark, TAND achieves state-of-the-art performance, surpassing both tissue-agnostic baselines and mask-supervised methods. Notably, our approach demonstrates remarkable improvements in tissue-dependent cell types such as epithelium, endothelium, and stroma. To the best of our knowledge, this is the first method to condition per-cell classification on learned tissue masks, offering a practical pathway to reduce annotation burden.
Authors: Ashlesha G. Sawant, Shreyash S. Kamble, Raj S. Kanade, Raunak N. Kanugo, Tanishq A. Kapse, Karan A. Bhapse
Abstract: One of the major causes of road accidents is driver fatigue that causes thousands of fatalities and injuries every year. This study shows development of a Driver Drowsiness Detection System meant to improve the safety of the road by alerting drivers who are showing signs of being drowsy. The system is based on a standard webcam that tracks the facial features of the driver with the main emphasis on the examination of eye movements that can be conducted with the help of the Eye Aspect Ratio (EAR) method. The Face Mesh by MediaPipe is a lightweight framework that can identify facial landmarks with high accuracy and efficiency, which is considered to be important in real time use. The system detects the moments of long eye shutdowns or a very low rate of blinking which are manifestations of drowsiness and alerts the driver through sound to get her attention back. This system achieves a high-performance and low-cost driver monitoring solution with the help of the computational power of OpenCV to process the image and the MediaPipe to identify faces. Test data experimental analyses indicate that the system is very accurate and responds quicker; this confirms that it can be a component of the current Advanced Driving Assistance System (ADAS).
Authors: Dimitrios Koutsianos, Ladislav Mosner, Yannis Panagakis, Themos Stafylakis
Abstract: Performance in face and speaker verification is largely driven by margin based softmax losses like CosFace and ArcFace. Recently introduced $\alpha$-divergence loss functions offer a compelling alternative, particularly for their ability to induce sparse solutions (when $\alpha>1$). However, integrating an angular margin-crucial for verification tasks-is not straightforward. We find this integration can be achieved in at least two distinct ways: via the reference measure (prior probabilities) or via the logits (unnormalized log-likelihoods). In this paper, we explore both pathways, deriving two novel margin-based $\alpha$-divergence losses: Q-Margin (margin in the reference measure) and A3M (margin in the logits). We identify and address a critical training instability in A3M-caused by the interplay of penalized logits and sparsity-with a simple yet effective prototype re-initialization strategy. Our methods achieve significant performance gains on the challenging IJB-B and IJB-C face verification benchmarks. We demonstrate similarly strong performance in speaker verification on VoxCeleb. Crucially, our models significantly outperform strong baselines at low false acceptance rates (FAR). This capability is crucial for practical high-security applications, such as banking authentication, when minimizing false authentications is paramount.
Authors: Shrenik Patel, Daivik Patel
Abstract: Long-form video question answering (VQA) overwhelms current vision-language models (VLMs) because attention and key-value (KV) caches grow with runtime, forcing either expensive inference or near-sighted sliding windows. We introduce CacheFlow, a training-free pipeline that pairs Dynamic Token Dropping (DTD) with a compressive long-term memory. DTD prunes per-patch tokens online via cosine similarity to the previous frame, and surviving tokens are packed into fixed-size blocks. This online, per-frame processing makes our approach fundamentally suited for live streaming VQA. As blocks are processed, each one's keys are summarized by a tiny recurrent encoder to form a retrieval index, while the block's full KV pairs are offloaded and later rehydrated for generation, preserving answer fidelity. At inference, a consensus-based retrieval mechanism retrieves only the Top-K most relevant blocks and attends over both the retrieved and local context for precise, long-range reasoning. CacheFlow is drop-in, architecture-agnostic, and requires no fine-tuning. Experiments on both offline and streaming VQA benchmarks demonstrate that CacheFlow outperforms current strong baselines, while processing up to 87% less tokens. Our dual approach enables VLMs to be both efficient and context-aware, paving the way for practical long-form video understanding.
Authors: Chunshi Wang, Junliang Ye, Yunhan Yang, Yang Li, Zizhuo Lin, Jun Zhu, Zhuo Chen, Yawei Luo, Chunchao Guo
Abstract: We introduce Part-X-MLLM, a native 3D multimodal large language model that unifies diverse 3D tasks by formulating them as programs in a structured, executable grammar. Given an RGB point cloud and a natural language prompt, our model autoregressively generates a single, coherent token sequence encoding part-level bounding boxes, semantic descriptions, and edit commands. This structured output serves as a versatile interface to drive downstream geometry-aware modules for part-based generation and editing. By decoupling the symbolic planning from the geometric synthesis, our approach allows any compatible geometry engine to be controlled through a single, language-native frontend. We pre-train a dual-encoder architecture to disentangle structure from semantics and instruction-tune the model on a large-scale, part-centric dataset. Experiments demonstrate that our model excels at producing high-quality, structured plans, enabling state-of-the-art performance in grounded Q\&A, compositional generation, and localized editing through one unified interface. Project page: https://chunshi.wang/Part-X-MLLM/
Authors: Ziang Cao, Fangzhou Hong, Zhaoxi Chen, Liang Pan, Ziwei Liu
Abstract: 3D modeling is shifting from static visual representations toward physical, articulated assets that can be directly used in simulation and interaction. However, most existing 3D generation methods overlook key physical and articulation properties, thereby limiting their utility in embodied AI. To bridge this gap, we introduce PhysX-Anything, the first simulation-ready physical 3D generative framework that, given a single in-the-wild image, produces high-quality sim-ready 3D assets with explicit geometry, articulation, and physical attributes. Specifically, we propose the first VLM-based physical 3D generative model, along with a new 3D representation that efficiently tokenizes geometry. It reduces the number of tokens by 193x, enabling explicit geometry learning within standard VLM token budgets without introducing any special tokens during fine-tuning and significantly improving generative quality. In addition, to overcome the limited diversity of existing physical 3D datasets, we construct a new dataset, PhysX-Mobility, which expands the object categories in prior physical 3D datasets by over 2x and includes more than 2K common real-world objects with rich physical annotations. Extensive experiments on PhysX-Mobility and in-the-wild images demonstrate that PhysX-Anything delivers strong generative performance and robust generalization. Furthermore, simulation-based experiments in a MuJoCo-style environment validate that our sim-ready assets can be directly used for contact-rich robotic policy learning. We believe PhysX-Anything can substantially empower a broad range of downstream applications, especially in embodied AI and physics-based simulation.
Authors: Dengyang Jiang, Dongyang Liu, Zanyi Wang, Qilong Wu, Xin Jin, David Liu, Zhen Li, Mengmeng Wang, Peng Gao, Harry Yang
Abstract: Distribution Matching Distillation (DMD) distills a pre-trained multi-step diffusion model to a few-step one to improve inference efficiency. However, the performance of the latter is often capped by the former. To circumvent this dilemma, we propose DMDR, a novel framework that combines Reinforcement Learning (RL) techniques into the distillation process. We show that for the RL of the few-step generator, the DMD loss itself is a more effective regularization compared to the traditional ones. In turn, RL can help to guide the mode coverage process in DMD more effectively. These allow us to unlock the capacity of the few-step generator by conducting distillation and RL simultaneously. Meanwhile, we design the dynamic distribution guidance and dynamic renoise sampling training strategies to improve the initial distillation process. The experiments demonstrate that DMDR can achieve leading visual quality, prompt coherence among few-step methods, and even exhibit performance that exceeds the multi-step teacher.
Authors: Henry Herzog, Favyen Bastani, Yawen Zhang, Gabriel Tseng, Joseph Redmon, Hadrien Sablon, Ryan Park, Jacob Morrison, Alexandra Buraczynski, Karen Farley, Joshua Hansen, Andrew Howe, Patrick Alan Johnson, Mark Otterlee, Ted Schmitt, Hunter Pitelka, Stephen Daspit, Rachel Ratner, Christopher Wilhelm, Sebastian Wood, Mike Jacobi, Hannah Kerner, Evan Shelhamer, Ali Farhadi, Ranjay Krishna, Patrick Beukema
Abstract: Earth observation data presents a unique challenge: it is spatial like images, sequential like video or text, and highly multimodal. We present OlmoEarth: a multimodal, spatio-temporal foundation model that employs a novel self-supervised learning formulation, masking strategy, and loss all designed for the Earth observation domain. OlmoEarth achieves state-of-the-art performance compared to 12 other foundation models across a variety of research benchmarks and real-world tasks from external partners. When evaluating embeddings OlmoEarth achieves the best performance on 15 out of 24 tasks, and with full fine-tuning it is the best on 19 of 29 tasks. We deploy OlmoEarth as the backbone of an end-to-end platform for data collection, labeling, training, and inference of Earth observation models. The OlmoEarth Platform puts frontier foundation models and powerful data management tools into the hands of non-profits and NGOs working to solve the world's biggest problems. OlmoEarth source code, training data, and pre-trained weights are available at $\href{https://github.com/allenai/olmoearth_pretrain}{\text{https://github.com/allenai/olmoearth_pretrain}}$.
URLs: https://github.com/allenai/olmoearth_pretrain, https://github.com/allenai/olmoearth_pretrain
Authors: Jiangnan Ye, Jiedong Zhuang, Lianrui Mu, Wenjie Zheng, Jiaqi Hu, Xingze Zou, Jing Wang, Haoji Hu
Abstract: We introduce GS-Light, an efficient, textual position-aware pipeline for text-guided relighting of 3D scenes represented via Gaussian Splatting (3DGS). GS-Light implements a training-free extension of a single-input diffusion model to handle multi-view inputs. Given a user prompt that may specify lighting direction, color, intensity, or reference objects, we employ a large vision-language model (LVLM) to parse the prompt into lighting priors. Using off-the-shelf estimators for geometry and semantics (depth, surface normals, and semantic segmentation), we fuse these lighting priors with view-geometry constraints to compute illumination maps and generate initial latent codes for each view. These meticulously derived init latents guide the diffusion model to generate relighting outputs that more accurately reflect user expectations, especially in terms of lighting direction. By feeding multi-view rendered images, along with the init latents, into our multi-view relighting model, we produce high-fidelity, artistically relit images. Finally, we fine-tune the 3DGS scene with the relit appearance to obtain a fully relit 3D scene. We evaluate GS-Light on both indoor and outdoor scenes, comparing it to state-of-the-art baselines including per-view relighting, video relighting, and scene editing methods. Using quantitative metrics (multi-view consistency, imaging quality, aesthetic score, semantic similarity, etc.) and qualitative assessment (user studies), GS-Light demonstrates consistent improvements over baselines. Code and assets will be made available upon publication.
Authors: Harold Haodong Chen, Disen Lan, Wen-Jie Shu, Qingyang Liu, Zihan Wang, Sirui Chen, Wenkai Cheng, Kanghao Chen, Hongfei Zhang, Zixin Zhang, Rongjin Guo, Yu Cheng, Ying-Cong Chen
Abstract: The rapid evolution of video generative models has shifted their focus from producing visually plausible outputs to tackling tasks requiring physical plausibility and logical consistency. However, despite recent breakthroughs such as Veo 3's chain-of-frames reasoning, it remains unclear whether these models can exhibit reasoning capabilities similar to large language models (LLMs). Existing benchmarks predominantly evaluate visual fidelity and temporal coherence, failing to capture higher-order reasoning abilities. To bridge this gap, we propose TiViBench, a hierarchical benchmark specifically designed to evaluate the reasoning capabilities of image-to-video (I2V) generation models. TiViBench systematically assesses reasoning across four dimensions: i) Structural Reasoning & Search, ii) Spatial & Visual Pattern Reasoning, iii) Symbolic & Logical Reasoning, and iv) Action Planning & Task Execution, spanning 24 diverse task scenarios across 3 difficulty levels. Through extensive evaluations, we show that commercial models (e.g., Sora 2, Veo 3.1) demonstrate stronger reasoning potential, while open-source models reveal untapped potential that remains hindered by limited training scale and data diversity. To further unlock this potential, we introduce VideoTPO, a simple yet effective test-time strategy inspired by preference optimization. By performing LLM self-analysis on generated candidates to identify strengths and weaknesses, VideoTPO significantly enhances reasoning performance without requiring additional training, data, or reward models. Together, TiViBench and VideoTPO pave the way for evaluating and advancing reasoning in video generation models, setting a foundation for future research in this emerging field.
Authors: Xincheng Shuai, Zhenyuan Qin, Henghui Ding, Dacheng Tao
Abstract: Recent advances in text-to-image (T2I) diffusion models have significantly improved semantic image editing, yet most methods fall short in performing 3D-aware object manipulation. In this work, we present FFSE, a 3D-aware autoregressive framework designed to enable intuitive, physically-consistent object editing directly on real-world images. Unlike previous approaches that either operate in image space or require slow and error-prone 3D reconstruction, FFSE models editing as a sequence of learned 3D transformations, allowing users to perform arbitrary manipulations, such as translation, scaling, and rotation, while preserving realistic background effects (e.g., shadows, reflections) and maintaining global scene consistency across multiple editing rounds. To support learning of multi-round 3D-aware object manipulation, we introduce 3DObjectEditor, a hybrid dataset constructed from simulated editing sequences across diverse objects and scenes, enabling effective training under multi-round and dynamic conditions. Extensive experiments show that the proposed FFSE significantly outperforms existing methods in both single-round and multi-round 3D-aware editing scenarios.
Authors: Junwei Yu, Trevor Darrell, XuDong Wang
Abstract: The Segment Anything Model (SAM) family has become a widely adopted vision foundation model, but its ability to control segmentation granularity remains limited. Users often need to refine results manually - by adding more prompts or selecting from pre-generated masks - to achieve the desired level of detail. This process can be ambiguous, as the same prompt may correspond to several plausible masks, and collecting dense annotations across all granularities is prohibitively expensive, making supervised solutions infeasible. To address this limitation, we introduce UnSAMv2, which enables segment anything at any granularity without human annotations. UnSAMv2 extends the divide-and-conquer strategy of UnSAM by discovering abundant mask-granularity pairs and introducing a novel granularity control embedding that enables precise, continuous control over segmentation scale. Remarkably, with only $6$K unlabeled images and $0.02\%$ additional parameters, UnSAMv2 substantially enhances SAM-2, achieving segment anything at any granularity across interactive, whole-image, and video segmentation tasks. Evaluated on over $11$ benchmarks, UnSAMv2 improves $\text{NoC}_{90}$ (5.69 $\rightarrow$ 4.75), 1-IoU (58.0 $\rightarrow$ 73.1), and $\text{AR}_{1000}$ (49.6 $\rightarrow$ 68.3), showing that small amounts of unlabeled data with a granularity-aware self-supervised learning method can unlock the potential of vision foundation models.
Authors: Hengrui Hu, Kaining Ying, Henghui Ding
Abstract: This work focuses on multi-shot semi-supervised video object segmentation (MVOS), which aims at segmenting the target object indicated by an initial mask throughout a video with multiple shots. The existing VOS methods mainly focus on single-shot videos and struggle with shot discontinuities, thereby limiting their real-world applicability. We propose a transition mimicking data augmentation strategy (TMA) which enables cross-shot generalization with single-shot data to alleviate the severe annotated multi-shot data sparsity, and the Segment Anything Across Shots (SAAS) model, which can detect and comprehend shot transitions effectively. To support evaluation and future study in MVOS, we introduce Cut-VOS, a new MVOS benchmark with dense mask annotations, diverse object categories, and high-frequency transitions. Extensive experiments on YouMVOS and Cut-VOS demonstrate that the proposed SAAS achieves state-of-the-art performance by effectively mimicking, understanding, and segmenting across complex transitions. The code and datasets are released at https://henghuiding.com/SAAS/.
Authors: Zhongang Cai, Ruisi Wang, Chenyang Gu, Fanyi Pu, Junxiang Xu, Yubo Wang, Wanqi Yin, Zhitao Yang, Chen Wei, Qingping Sun, Tongxi Zhou, Jiaqi Li, Hui En Pang, Oscar Qian, Yukun Wei, Zhiqian Lin, Xuanke Shi, Kewang Deng, Xiaoyang Han, Zukai Chen, Xiangyu Fan, Hanming Deng, Lewei Lu, Liang Pan, Bo Li, Ziwei Liu, Quan Wang, Dahua Lin, Lei Yang
Abstract: Despite remarkable progress, multimodal foundation models still exhibit surprising deficiencies in spatial intelligence. In this work, we explore scaling up multimodal foundation models to cultivate spatial intelligence within the SenseNova-SI family, built upon established multimodal foundations including visual understanding models (i.e., Qwen3-VL and InternVL3) and unified understanding and generation models (i.e., Bagel). We take a principled approach to constructing high-performing and robust spatial intelligence by systematically curating SenseNova-SI-8M: eight million diverse data samples under a rigorous taxonomy of spatial capabilities. SenseNova-SI demonstrates unprecedented performance across a broad range of spatial intelligence benchmarks: 68.7% on VSI-Bench, 43.3% on MMSI, 85.6% on MindCube, 54.6% on ViewSpatial, and 50.1% on SITE, while maintaining strong general multimodal understanding (e.g., 84.9% on MMBench-En). More importantly, we analyze the impact of data scaling, discuss early signs of emergent generalization capabilities enabled by diverse data training, analyze the risk of overfitting and language shortcuts, present a preliminary study on spatial chain-of-thought reasoning, and validate the potential downstream application. SenseNova-SI is an ongoing project, and this report will be updated continuously. All newly trained multimodal foundation models are publicly released to facilitate further research in this direction.
Authors: Tianhong Li, Kaiming He
Abstract: Today's denoising diffusion models do not "denoise" in the classical sense, i.e., they do not directly predict clean images. Rather, the neural networks predict noise or a noised quantity. In this paper, we suggest that predicting clean data and predicting noised quantities are fundamentally different. According to the manifold assumption, natural data should lie on a low-dimensional manifold, whereas noised quantities do not. With this assumption, we advocate for models that directly predict clean data, which allows apparently under-capacity networks to operate effectively in very high-dimensional spaces. We show that simple, large-patch Transformers on pixels can be strong generative models: using no tokenizer, no pre-training, and no extra loss. Our approach is conceptually nothing more than "$\textbf{Just image Transformers}$", or $\textbf{JiT}$, as we call it. We report competitive results using JiT with large patch sizes of 16 and 32 on ImageNet at resolutions of 256 and 512, where predicting high-dimensional noised quantities can fail catastrophically. With our networks mapping back to the basics of the manifold, our research goes back to basics and pursues a self-contained paradigm for Transformer-based diffusion on raw natural data.
Authors: Michikuni Eguchi, Takekazu Kitagishi, Yuichi Hiroi, Takefumi Hiraki
Abstract: The tactile sensation of clothing is critical to wearer comfort. To reveal physical properties that make clothing comfortable, systematic collection of tactile data during sliding motion is required. We propose a robotic arm-based system for collecting tactile data from intact garments. The system performs stroking measurements with a simulated fingertip while precisely controlling speed and direction, enabling creation of motion-labeled, multimodal tactile databases. Machine learning evaluation showed that including motion-related parameters improved identification accuracy for audio and acceleration data, demonstrating the efficacy of motion-related labels for characterizing clothing tactile sensation. This system provides a scalable, non-destructive method for capturing tactile data of clothing, contributing to future studies on fabric perception and reproduction.
Authors: Jie Fan, Francesco Visentin, Barbara Mazzolai, Emanuela Del Dottore
Abstract: Tendrils coil their shape to anchor the plant to supporting structures, allowing vertical growth toward light. Although climbing plants have been studied for a long time, extracting information regarding the relationship between the temporal shape change, the event that triggers it, and the contact location is still challenging. To help build this relation, we propose an image-based method by which it is possible to analyze shape changes over time in tendrils when mechano-stimulated in different portions of their body. We employ a geometric approach using a 3D Piece-Wise Clothoid-based model to reconstruct the configuration taken by a tendril after mechanical rubbing. The reconstruction shows high robustness and reliability with an accuracy of R2 > 0.99. This method demonstrates distinct advantages over deep learning-based approaches, including reduced data requirements, lower computational costs, and interpretability. Our analysis reveals higher responsiveness in the apical segment of tendrils, which might correspond to higher sensitivity and tissue flexibility in that region of the organs. Our study provides a methodology for gaining new insights into plant biomechanics and offers a foundation for designing and developing novel intelligent robotic systems inspired by climbing plants.
Authors: Jiantang Huang
Abstract: Basketball broadcast footage is traditionally captured at 30-60 fps, limiting viewers' ability to appreciate rapid plays such as dunks and crossovers. We present a real-time slow-motion synthesis system that produces high-quality basketball-specific interpolated frames by fine-tuning the recent Real-Time Intermediate Flow Estimation (RIFE) network on the SportsSloMo dataset. Our pipeline isolates the basketball subset of SportsSloMo, extracts training triplets, and fine-tunes RIFE with human-aware random cropping. We compare the resulting model against Super SloMo and the baseline RIFE model using Peak Signal-to-Noise Ratio (PSNR) and Structural Similarity (SSIM) on held-out clips. The fine-tuned RIFE attains a mean PSNR of 34.3 dB and SSIM of 0.949, outperforming Super SloMo by 2.1 dB and the baseline RIFE by 1.3 dB. A lightweight Gradio interface demonstrates end-to-end 4x slow-motion generation on a single RTX 4070 Ti Super at approximately 30 fps. These results indicate that task-specific adaptation is crucial for sports slow-motion, and that RIFE provides an attractive accuracy-speed trade-off for consumer applications.
Authors: Mingyu Sung, Suhwan Im, Vikas Palakonda, Jae-Mo Kang
Abstract: Split computing distributes deep neural network inference between resource-constrained edge devices and cloud servers but faces significant communication bottlenecks when transmitting intermediate features. To this end, in this paper, we propose a novel lightweight compression framework that leverages Range Asymmetric Numeral Systems (rANS) encoding with asymmetric integer quantization and sparse tensor representation to reduce transmission overhead dramatically. Specifically, our approach combines asymmetric integer quantization with a sparse representation technique, eliminating the need for complex probability modeling or network modifications. The key contributions include: (1) a distribution-agnostic compression pipeline that exploits inherent tensor sparsity to achieve bandwidth reduction with minimal computational overhead; (2) an approximate theoretical model that optimizes tensor reshaping dimensions to maximize compression efficiency; and (3) a GPU-accelerated implementation with sub-millisecond encoding/decoding latency. Extensive evaluations across diverse neural architectures (ResNet, VGG16, MobileNetV2, SwinT, DenseNet121, EfficientNetB0) demonstrate that the proposed framework consistently maintains near-baseline accuracy across CIFAR100 and ImageNet benchmarks. Moreover, we validated the framework's effectiveness on advanced natural language processing tasks by employing Llama2 7B and 13B on standard benchmarks such as MMLU, HellaSwag, ARC, PIQA, Winogrande, BoolQ, and OpenBookQA, demonstrating its broad applicability beyond computer vision. Furthermore, this method addresses a fundamental bottleneck in deploying sophisticated artificial intelligence systems in bandwidth-constrained environments without compromising model performance.
Authors: Hanchen David Wang, Siwoo Bae, Zirong Chen, Meiyi Ma
Abstract: Artificial intelligence systems in critical fields like autonomous driving and medical imaging analysis often continually learn new tasks using a shared stream of input data. For instance, after learning to detect traffic signs, a model may later need to learn to classify traffic lights or different types of vehicles using the same camera feed. This scenario introduces a challenging setting we term Continual Multitask Learning (CMTL), where a model sequentially learns new tasks on an underlying data distribution without forgetting previously learned abilities. Existing continual learning methods often fail in this setting because they learn fragmented, task-specific features that interfere with one another. To address this, we introduce Learning with Preserving (LwP), a novel framework that shifts the focus from preserving task outputs to maintaining the geometric structure of the shared representation space. The core of LwP is a Dynamically Weighted Distance Preservation (DWDP) loss that prevents representation drift by regularizing the pairwise distances between latent data representations. This mechanism of preserving the underlying geometric structure allows the model to retain implicit knowledge and support diverse tasks without requiring a replay buffer, making it suitable for privacy-conscious applications. Extensive evaluations on time-series and image benchmarks show that LwP not only mitigates catastrophic forgetting but also consistently outperforms state-of-the-art baselines in CMTL tasks. Notably, our method shows superior robustness to distribution shifts and is the only approach to surpass the strong single-task learning baseline, underscoring its effectiveness for real-world dynamic environments.
Authors: Zhehao Xu, Lok Ming Lui
Abstract: Free-boundary diffeomorphism optimization is a core ingredient in the surface mapping problem but remains notoriously difficult because the boundary is unconstrained and local bijectivity must be preserved under large deformation. Numerical Least-Squares Quasiconformal (LSQC) theory, with its provable existence, uniqueness, similarity-invariance and resolution-independence, offers an elegant mathematical remedy. However, the conventional numerical algorithm requires landmark conditioning, and cannot be applied into gradient-based optimization. We propose a neural surrogate, the Spectral Beltrami Network (SBN), that embeds LSQC energy into a multiscale mesh-spectral architecture. Next, we propose the SBN guided optimization framework SBN-Opt which optimizes free-boundary diffeomorphism for the problem, with local geometric distortion explicitly controllable. Extensive experiments on density-equalizing maps and inconsistent surface registration demonstrate our SBN-Opt's superiority over traditional numerical algorithms.
Authors: Udaya Bhasker Cheerala, Varun Teja Chirukuri, Venkata Akhil Kumar Gummadi, Jintu Moni Bhuyan, Praveen Damacharla
Abstract: Wildfires pose a significant global threat to ecosystems worldwide, with California experiencing recurring fires due to various factors, including climate, topographical features, vegetation patterns, and human activities. This study aims to develop a comprehensive wildfire risk map for California by applying the random forest (RF) algorithm, augmented with Explainable Artificial Intelligence (XAI) through Shapley Additive exPlanations (SHAP), to interpret model predictions. Model performance was assessed using both spatial and temporal validation strategies. The RF model demonstrated strong predictive performance, achieving near-perfect discrimination for grasslands (AUC = 0.996) and forests (AUC = 0.997). Spatial cross-validation revealed moderate transferability, yielding ROC-AUC values of 0.6155 for forests and 0.5416 for grasslands. In contrast, temporal split validation showed enhanced generalization, especially for forests (ROC-AUC = 0.6615, PR-AUC = 0.8423). SHAP-based XAI analysis identified key ecosystem-specific drivers: soil organic carbon, tree cover, and Normalized Difference Vegetation Index (NDVI) emerged as the most influential in forests, whereas Land Surface Temperature (LST), elevation, and vegetation health indices were dominant in grasslands. District-level classification revealed that Central Valley and Northern Buttes districts had the highest concentration of high-risk grasslands, while Northern Buttes and North Coast Redwoods dominated forested high-risk areas. This RF-SHAP framework offers a robust, comprehensible, and adaptable method for assessing wildfire risks, enabling informed decisions and creating targeted strategies to mitigate dangers.
Authors: Penghui Niu, Jiashuai She, Taotao Cai, Yajuan Zhang, Ping Zhang, Junhua Gu, Jianxin Li
Abstract: Ground-based cloud image segmentation is a critical research domain for photovoltaic power forecasting. Current deep learning approaches primarily focus on encoder-decoder architectural refinements. However, existing methodologies exhibit several limitations:(1)they rely on dilated convolutions for multi-scale context extraction, lacking the partial feature effectiveness and interoperability of inter-channel;(2)attention-based feature enhancement implementations neglect accuracy-throughput balance; and (3)the decoder modifications fail to establish global interdependencies among hierarchical local features, limiting inference efficiency. To address these challenges, we propose MPCM-Net, a Multi-scale network that integrates Partial attention Convolutions with Mamba architectures to enhance segmentation accuracy and computational efficiency. Specifically, the encoder incorporates MPAC, which comprises:(1)a MPC block with ParCM and ParSM that enables global spatial interaction across multi-scale cloud formations, and (2)a MPA block combining ParAM and ParSM to extract discriminative features with reduced computational complexity. On the decoder side, a M2B is employed to mitigate contextual loss through a SSHD that maintains linear complexity while enabling deep feature aggregation across spatial and scale dimensions. As a key contribution to the community, we also introduce and release a dataset CSRC, which is a clear-label, fine-grained segmentation benchmark designed to overcome the critical limitations of existing public datasets. Extensive experiments on CSRC demonstrate the superior performance of MPCM-Net over state-of-the-art methods, achieving an optimal balance between segmentation accuracy and inference speed. The dataset and source code will be available at https://github.com/she1110/CSRC.
Authors: Longhua Li, Lei Qi, Xin Geng
Abstract: Training and deploying multiple vision transformer (ViT) models for different resource constraints is costly and inefficient. To address this, we propose transforming a pre-trained ViT into a stratified knowledge-density super-network, where knowledge is hierarchically organized across weights. This enables flexible extraction of sub-networks that retain maximal knowledge for varying model sizes. We introduce \textbf{W}eighted \textbf{P}CA for \textbf{A}ttention \textbf{C}ontraction (WPAC), which concentrates knowledge into a compact set of critical weights. WPAC applies token-wise weighted principal component analysis to intermediate features and injects the resulting transformation and inverse matrices into adjacent layers, preserving the original network function while enhancing knowledge compactness. To further promote stratified knowledge organization, we propose \textbf{P}rogressive \textbf{I}mportance-\textbf{A}ware \textbf{D}ropout (PIAD). PIAD progressively evaluates the importance of weight groups, updates an importance-aware dropout list, and trains the super-network under this dropout regime to promote knowledge stratification. Experiments demonstrate that WPAC outperforms existing pruning criteria in knowledge concentration, and the combination with PIAD offers a strong alternative to state-of-the-art model compression and model expansion methods.
Authors: Aihua Zhu, Rui Su, Qinglin Zhao, Li Feng, Meng Shen, Shibo He
Abstract: Diffusion probabilistic models have set a new standard for generative fidelity but are hindered by a slow iterative sampling process. A powerful training-free strategy to accelerate this process is Schedule Optimization, which aims to find an optimal distribution of timesteps for a fixed and small Number of Function Evaluations (NFE) to maximize sample quality. To this end, a successful schedule optimization method must adhere to four core principles: effectiveness, adaptivity, practical robustness, and computational efficiency. However, existing paradigms struggle to satisfy these principles simultaneously, motivating the need for a more advanced solution. To overcome these limitations, we propose the Hierarchical-Schedule-Optimizer (HSO), a novel and efficient bi-level optimization framework. HSO reframes the search for a globally optimal schedule into a more tractable problem by iteratively alternating between two synergistic levels: an upper-level global search for an optimal initialization strategy and a lower-level local optimization for schedule refinement. This process is guided by two key innovations: the Midpoint Error Proxy (MEP), a solver-agnostic and numerically stable objective for effective local optimization, and the Spacing-Penalized Fitness (SPF) function, which ensures practical robustness by penalizing pathologically close timesteps. Extensive experiments show that HSO sets a new state-of-the-art for training-free sampling in the extremely low-NFE regime. For instance, with an NFE of just 5, HSO achieves a remarkable FID of 11.94 on LAION-Aesthetics with Stable Diffusion v2.1. Crucially, this level of performance is attained not through costly retraining, but with a one-time optimization cost of less than 8 seconds, presenting a highly practical and efficient paradigm for diffusion model acceleration.
Authors: Fei Song, Yi Li, Rui Wang, Jiahuan Zhou, Changwen Zheng, Jiangmeng Li
Abstract: Test-time prompt tuning for vision-language models has demonstrated impressive generalization capabilities under zero-shot settings. However, tuning the learnable prompts solely based on unlabeled test data may induce prompt optimization bias, ultimately leading to suboptimal performance on downstream tasks. In this work, we analyze the underlying causes of prompt optimization bias from both the model and data perspectives. In terms of the model, the entropy minimization objective typically focuses on reducing the entropy of model predictions while overlooking their correctness. This can result in overconfident yet incorrect outputs, thereby compromising the quality of prompt optimization. On the data side, prompts affected by optimization bias can introduce misalignment between visual and textual modalities, which further aggravates the prompt optimization bias. To this end, we propose a Doubly Debiased Test-Time Prompt Tuning method. Specifically, we first introduce a dynamic retrieval-augmented modulation module that retrieves high-confidence knowledge from a dynamic knowledge base using the test image feature as a query, and uses the retrieved knowledge to modulate the predictions. Guided by the refined predictions, we further develop a reliability-aware prompt optimization module that incorporates a confidence-based weighted ensemble and cross-modal consistency distillation to impose regularization constraints during prompt tuning. Extensive experiments across 15 benchmark datasets involving both natural distribution shifts and cross-datasets generalization demonstrate that our method outperforms baselines, validating its effectiveness in mitigating prompt optimization bias.
Authors: Jiayin Zhu, Linlin Yang, Yicong Li, Angela Yao
Abstract: Optimization-based text-to-3D methods distill guidance from 2D generative models via Score Distillation Sampling (SDS), but implicitly treat this guidance as static. This work shows that ignoring source dynamics yields inconsistent trajectories that suppress or merge semantic cues, leading to "semantic over-smoothing" artifacts. As such, we reformulate text-to-3D optimization as mapping a dynamically evolving source distribution to a fixed target distribution. We cast the problem into a dual-conditioned latent space, conditioned on both the text prompt and the intermediately rendered image. Given this joint setup, we observe that the image condition naturally anchors the current source distribution. Building on this insight, we introduce AnchorDS, an improved score distillation mechanism that provides state-anchored guidance with image conditions and stabilizes generation. We further penalize erroneous source estimates and design a lightweight filter strategy and fine-tuning strategy that refines the anchor with negligible overhead. AnchorDS produces finer-grained detail, more natural colours, and stronger semantic consistency, particularly for complex prompts, while maintaining efficiency. Extensive experiments show that our method surpasses previous methods in both quality and efficiency.
Authors: Xin Zhao, Xiaojun Chen, Bingshan Liu, Zeyao Liu, Zhendong Zhao, Xiaoyan Gu
Abstract: Generative vision-language models like Stable Diffusion demonstrate remarkable capabilities in creative media synthesis, but they also pose substantial risks of producing unsafe, offensive, or culturally inappropriate content when prompted adversarially. Current defenses struggle to align outputs with human values without sacrificing generation quality or incurring high costs. To address these challenges, we introduce VALOR (Value-Aligned LLM-Overseen Rewriter), a modular, zero-shot agentic framework for safer and more helpful text-to-image generation. VALOR integrates layered prompt analysis with human-aligned value reasoning: a multi-level NSFW detector filters lexical and semantic risks; a cultural value alignment module identifies violations of social norms, legality, and representational ethics; and an intention disambiguator detects subtle or indirect unsafe implications. When unsafe content is detected, prompts are selectively rewritten by a large language model under dynamic, role-specific instructions designed to preserve user intent while enforcing alignment. If the generated image still fails a safety check, VALOR optionally performs a stylistic regeneration to steer the output toward a safer visual domain without altering core semantics. Experiments across adversarial, ambiguous, and value-sensitive prompts show that VALOR significantly reduces unsafe outputs by up to 100.00% while preserving prompt usefulness and creativity. These results highlight VALOR as a scalable and effective approach for deploying safe, aligned, and helpful image generation systems in open-world settings.
Authors: Xun Shao, Aoba Otani, Yuto Hirasuka, Runji Cai, Seng W. Loke
Abstract: This position paper envisions a next-generation elderly monitoring system that moves beyond fall detection toward the broader goal of Activities of Daily Living (ADL) recognition. Our ultimate aim is to design privacy-preserving, edge-deployed, and federated AI systems that can robustly detect and understand daily routines, supporting independence and dignity in aging societies. At present, ADL-specific datasets are still under collection. As a preliminary step, we demonstrate feasibility through experiments using the SISFall dataset and its GAN-augmented variants, treating fall detection as a proxy task. We report initial results on federated learning with non-IID conditions, and embedded deployment on Jetson Orin Nano devices. We then outline open challenges such as domain shift, data scarcity, and privacy risks, and propose directions toward full ADL monitoring in smart-room environments. This work highlights the transition from single-task detection to comprehensive daily activity recognition, providing both early evidence and a roadmap for sustainable and human-centered elderly care AI.
Authors: Matvey Skripkin, Elizaveta Goncharova, Andrey Kuznetsov
Abstract: We present a lightweight yet effective pipeline for training vision-language models to solve math problems by rendering LaTeX encoded equations into images and pairing them with structured chain-of-thought prompts. This simple text-to-vision augmentation enables compact multimodal architectures to achieve state-of-the-art reasoning accuracy. Through systematic ablations, we find that rendering fidelity and prompt design are the primary drivers of performance. Despite its simplicity, our approach consistently matches or surpasses both open-source and proprietary math-focused vision-language solvers on widely used benchmarks, while preserving broad general-domain competence - showing gains on tasks such as MMMU, ChartQA, and DocVQA of up to 20%.
Authors: Arya Narang
Abstract: This paper determines the extent to which short textual inputs (in this case, names of dishes) can improve calorie estimation compared to an image-only baseline model and whether any improvements are statistically significant. Utilizes the TensorFlow library and the Nutrition5k dataset (curated by Google) to train both an image-only CNN and multimodal CNN that accepts both text and an image as input. The MAE of calorie estimations was reduced by 1.06 kcal from 84.76 kcal to 83.70 kcal (1.25% improvement) when using the multimodal model.
Authors: Julia Peters, Karin Mora, Miguel D. Mahecha, Chaonan Ji, David Montero, Clemens Mosig, Guido Kraemer
Abstract: Earth observation (EO) foundation models have emerged as an effective approach to derive latent representations of the Earth system from various remote sensing sensors. These models produce embeddings that can be used as analysis-ready datasets, enabling the modelling of ecosystem dynamics without extensive sensor-specific preprocessing. However, existing models typically operate at fixed spatial or temporal scales, limiting their use for ecological analyses that require both fine spatial detail and high temporal fidelity. To overcome these limitations, we propose a representation learning framework that integrates different EO modalities into a unified feature space at high spatio-temporal resolution. We introduce the framework using Sentinel-1 and Sentinel-2 data as representative modalities. Our approach produces a latent space at native 10 m resolution and the temporal frequency of cloud-free Sentinel-2 acquisitions. Each sensor is first modeled independently to capture its sensor-specific characteristics. Their representations are then combined into a shared model. This two-stage design enables modality-specific optimisation and easy extension to new sensors, retaining pretrained encoders while retraining only fusion layers. This enables the model to capture complementary remote sensing data and to preserve coherence across space and time. Qualitative analyses reveal that the learned embeddings exhibit high spatial and semantic consistency across heterogeneous landscapes. Quantitative evaluation in modelling Gross Primary Production reveals that they encode ecologically meaningful patterns and retain sufficient temporal fidelity to support fine-scale analyses. Overall, the proposed framework provides a flexible, analysis-ready representation learning approach for environmental applications requiring diverse spatial and temporal resolutions.
Authors: Yunkai Yu, Yingying Wang, Rong Zheng
Abstract: The Internet of Things (IoT) sensors have been widely employed to capture human locomotions to enable applications such as activity recognition, human pose estimation, and fall detection. Motion capture (MoCap) systems are frequently used to generate ground truth annotations for human poses when training models with data from wearable or ambient sensors, and have been shown to be effective to synthesize data in these modalities. However, the representation of older adults, an increasingly important demographic in healthcare, in existing MoCap locomotion datasets has not been thoroughly examined. This work surveyed 41 publicly available datasets, identifying eight that include older adult motions and four that contain motions performed by younger actors annotated as old style. Older adults represent a small portion of participants overall, and few datasets provide full-body motion data for this group. To assess the fidelity of old-style walking motions, quantitative metrics are introduced, defining high fidelity as the ability to capture age-related differences relative to normative walking. Using gait parameters that are age-sensitive, robust to noise, and resilient to data scarcity, we found that old-style walking motions often exhibit overly controlled patterns and fail to faithfully characterize aging. These findings highlight the need for improved representation of older adults in motion datasets and establish a method to quantitatively evaluate the quality of old-style walking motions.
Authors: Soumyendu Sarkar, Antonio Guillen-Perez, Zachariah J Carmichael, Avisek Naug, Refik Mert Cam, Vineet Gundecha, Ashwin Ramesh Babu, Sahand Ghorbanpour, Ricardo Luna Gutierrez
Abstract: Reducing energy consumption and carbon emissions in data centers by enabling real-time temperature prediction is critical for sustainability and operational efficiency. Achieving this requires accurate modeling of the 3D temperature field to capture airflow dynamics and thermal interactions under varying operating conditions. Traditional thermal CFD solvers, while accurate, are computationally expensive and require expert-crafted meshes and boundary conditions, making them impractical for real-time use. To address these limitations, we develop a vision-based surrogate modeling framework that operates directly on a 3D voxelized representation of the data center, incorporating server workloads, fan speeds, and HVAC temperature set points. We evaluate multiple architectures, including 3D CNN U-Net variants, a 3D Fourier Neural Operator, and 3D vision transformers, to map these thermal inputs to high-fidelity heat maps. Our results show that the surrogate models generalize across data center configurations and achieve up to 20,000x speedup (hundreds of milliseconds vs. hours). This fast and accurate estimation of hot spots and temperature distribution enables real-time cooling control and workload redistribution, leading to substantial energy savings (7\%) and reduced carbon footprint.
Authors: Tongda Xu
Abstract: Many recent works utilize denoising score matching to optimize the conditional input of diffusion models. In this workshop paper, we demonstrate that such optimization breaks the equivalence between denoising score matching and exact score matching. Furthermore, we show that this bias leads to higher score norm. Additionally, we observe a similar bias when optimizing the data distribution using a pre-trained diffusion model. Finally, we discuss the wide range of works across different domains that are affected by this bias, including MAR for auto-regressive generation, PerCo for image compression, and DreamFusion for text to 3D generation.
Authors: Mehdi Khaleghi, Nastaran Khaleghi, Sobhan Sheykhivand, Sebelan Danishvar
Abstract: Systematic logistics, conveyance amenities and facilities as well as warehousing information play a key role in fostering profitable development in a supply chain. The aim of transformation in industries is the improvement of the resiliency regarding the supply chain. The resiliency policies are required for companies to affect the collaboration with logistics service providers positively. The decrement of air pollutant emissions is a persistent advantage of the efficient management of logistics and transportation in supply chain. The management of shipment type is a significant factor in analyzing the sustainability of logistics and supply chain. An automatic approach to predict the shipment type, logistics delay and traffic status are required to improve the efficiency of the supply chain management. A hybrid graphsage network (H-GSN) is proposed in this paper for multi-task purpose of logistics management in a supply chain. The shipment type, shipment status, traffic status, logistics ID and logistics delay are the objectives in this article regarding three different databases including DataCo, Shipping and Smart Logistcis available on Kaggle as supply chain logistics databases. The average accuracy of 97.8% and 100% are acquired for 10 kinds of logistics ID and 3 types of traffic status prediction in Smart Logistics dataset. The average accuracy of 98.7% and 99.4% are obtained for shipment type prediction in DataCo and logistics delay in Shipping database, respectively. The evaluation metrics for different logistics scenarios confirm the efficiency of the proposed method to improve the resilience and sustainability of the supply chain.
Authors: Vinit Mehta, Charu Sharma, Karthick Thiyagarajan
Abstract: With the rapid advancement of artificial intelligence and robotics, the integration of Large Language Models (LLMs) with 3D vision is emerging as a transformative approach to enhancing robotic sensing technologies. This convergence enables machines to perceive, reason and interact with complex environments through natural language and spatial understanding, bridging the gap between linguistic intelligence and spatial perception. This review provides a comprehensive analysis of state-of-the-art methodologies, applications and challenges at the intersection of LLMs and 3D vision, with a focus on next-generation robotic sensing technologies. We first introduce the foundational principles of LLMs and 3D data representations, followed by an in-depth examination of 3D sensing technologies critical for robotics. The review then explores key advancements in scene understanding, text-to-3D generation, object grounding and embodied agents, highlighting cutting-edge techniques such as zero-shot 3D segmentation, dynamic scene synthesis and language-guided manipulation. Furthermore, we discuss multimodal LLMs that integrate 3D data with touch, auditory and thermal inputs, enhancing environmental comprehension and robotic decision-making. To support future research, we catalog benchmark datasets and evaluation metrics tailored for 3D-language and vision tasks. Finally, we identify key challenges and future research directions, including adaptive model architectures, enhanced cross-modal alignment and real-time processing capabilities, which pave the way for more intelligent, context-aware and autonomous robotic sensing systems.
Authors: Vlad Rakhlin, Amir Jevnisek, Shai Avidan
Abstract: ReLU activations are the main bottleneck in Private Inference that is based on ResNet networks. This is because they incur significant inference latency. Reducing ReLU count is a discrete optimization problem, and there are two common ways to approach it. Most current state-of-the-art methods are based on a smooth approximation that jointly optimizes network accuracy and ReLU budget at once. However, the last hard thresholding step of the optimization usually introduces a large performance loss. We take an alternative approach that works directly in the discrete domain by leveraging Coordinate Descent as our optimization framework. In contrast to previous methods, this yields a sparse solution by design. We demonstrate, through extensive experiments, that our method is State of the Art on common benchmarks.
Authors: Sultan Hassan, Sambatra Andrianomena, Benjamin D. Wandelt
Abstract: Systematics contaminate observables, leading to distribution shifts relative to theoretically simulated signals-posing a major challenge for using pre-trained models to label such observables. Since systematics are often poorly understood and difficult to model, removing them directly and entirely may not be feasible. To address this challenge, we propose a novel method that aligns learned features between in-distribution (ID) and out-of-distribution (OOD) samples by optimizing a feature-alignment loss on the representations extracted from a pre-trained ID model. We first experimentally validate the method on the MNIST dataset using possible alignment losses, including mean squared error and optimal transport, and subsequently apply it to large-scale maps of neutral hydrogen. Our results show that optimal transport is particularly effective at aligning OOD features when parity between ID and OOD samples is unknown, even with limited data-mimicking real-world conditions in extracting information from large-scale surveys. Our code is available at https://github.com/sultan-hassan/feature-alignment-for-OOD-generalization.
URLs: https://github.com/sultan-hassan/feature-alignment-for-OOD-generalization.
Authors: Wenhao Zhou, Hao Zheng, Rong Zhao
Abstract: Large Vision-Language Models (LVLMs) typically align visual features from an encoder with a pre-trained Large Language Model (LLM). However, this makes the visual perception module a bottleneck, which constrains the overall capabilities of LVLMs. Conventional evaluation benchmarks, while rich in visual semantics, often contain unavoidable local shortcuts that can lead to an overestimation of models' perceptual abilities. Here, we introduce TopoPerception, a benchmark that leverages topological properties to rigorously evaluate the global visual perception capabilities of LVLMs across various granularities. Since topology depends on the global structure of an image and is invariant to local features, TopoPerception enables a shortcut-free assessment of global perception, fundamentally distinguishing it from semantically rich tasks. We evaluate state-of-the-art models on TopoPerception and find that even at the coarsest perceptual granularity, all models perform no better than random chance, indicating a profound inability to perceive global visual features. Notably, a consistent trend emerge within model families: more powerful models with stronger reasoning capabilities exhibit lower accuracy. This suggests that merely scaling up models is insufficient to address this deficit and may even exacerbate it. Progress may require new training paradigms or architectures. TopoPerception not only exposes a critical bottleneck in current LVLMs but also offers a lens and direction for improving their global visual perception. The data and code are publicly available at: https://github.com/Wenhao-Zhou/TopoPerception.
Authors: David Montero, Miguel D. Mahecha, Francesco Martinuzzi, C\'esar Aybar, Anne Klosterhalfen, Alexander Knohl, Jes\'us Anaya, Clemens Mosig, Sebastian Wieneke
Abstract: Monitoring the spatiotemporal dynamics of forest CO$_2$ uptake (Gross Primary Production, GPP), remains a central challenge in terrestrial ecosystem research. While Eddy Covariance (EC) towers provide high-frequency estimates, their limited spatial coverage constrains large-scale assessments. Remote sensing offers a scalable alternative, yet most approaches rely on single-sensor spectral indices and statistical models that are often unable to capture the complex temporal dynamics of GPP. Recent advances in deep learning (DL) and data fusion offer new opportunities to better represent the temporal dynamics of vegetation processes, but comparative evaluations of state-of-the-art DL models for multimodal GPP prediction remain scarce. Here, we explore the performance of two representative models for predicting GPP: 1) GPT-2, a transformer architecture, and 2) Long Short-Term Memory (LSTM), a recurrent neural network, using multivariate inputs. Overall, both achieve similar accuracy. But, while LSTM performs better overall, GPT-2 excels during extreme events. Analysis of temporal context length further reveals that LSTM attains similar accuracy using substantially shorter input windows than GPT-2, highlighting an accuracy-efficiency trade-off between the two architectures. Feature importance analysis reveals radiation as the dominant predictor, followed by Sentinel-2, MODIS land surface temperature, and Sentinel-1 contributions. Our results demonstrate how model architecture, context length, and multimodal inputs jointly determine performance in GPP prediction, guiding future developments of DL frameworks for monitoring terrestrial carbon dynamics.
Authors: Xi Li, Nicholas Matsumoto, Ujjwal Pasupulety, Atharva Deo, Cherine Yang, Jay Moran, Miguel E. Hernandez, Peter Wager, Jasmine Lin, Jeanine Kim, Alvin C. Goh, Christian Wagner, Geoffrey A. Sonn, Andrew J. Hung
Abstract: Fine-grained analysis of intraoperative behavior and its impact on patient outcomes remain a longstanding challenge. We present Frame-to-Outcome (F2O), an end-to-end system that translates tissue dissection videos into gesture sequences and uncovers patterns associated with postoperative outcomes. Leveraging transformer-based spatial and temporal modeling and frame-wise classification, F2O robustly detects consecutive short (~2 seconds) gestures in the nerve-sparing step of robot-assisted radical prostatectomy (AUC: 0.80 frame-level; 0.81 video-level). F2O-derived features (gesture frequency, duration, and transitions) predicted postoperative outcomes with accuracy comparable to human annotations (0.79 vs. 0.75; overlapping 95% CI). Across 25 shared features, effect size directions were concordant with small differences (~ 0.07), and strong correlation (r = 0.96, p < 1e-14). F2O also captured key patterns linked to erectile function recovery, including prolonged tissue peeling and reduced energy use. By enabling automatic interpretable assessment, F2O establishes a foundation for data-driven surgical feedback and prospective clinical decision support.
Authors: Tianyu Xu, Jihan Li, Penghe Zu, Pranav Sahay, Maruchi Kim, Jack Obeng-Marnu, Farley Miller, Xun Qian, Katrina Passarella, Mahitha Rachumalla, Rajeev Nongpiur, D. Shin
Abstract: In Extended Reality (XR), rendering sound that accurately simulates real-world acoustics is pivotal in creating lifelike and believable virtual experiences. However, existing XR spatial audio rendering methods often struggle with real-time adaptation to diverse physical scenes, causing a sensory mismatch between visual and auditory cues that disrupts user immersion. To address this, we introduce SAMOSA, a novel on-device system that renders spatially accurate sound by dynamically adapting to its physical environment. SAMOSA leverages a synergistic multimodal scene representation by fusing real-time estimations of room geometry, surface materials, and semantic-driven acoustic context. This rich representation then enables efficient acoustic calibration via scene priors, allowing the system to synthesize a highly realistic Room Impulse Response (RIR). We validate our system through technical evaluation using acoustic metrics for RIR synthesis across various room configurations and sound types, alongside an expert evaluation (N=12). Evaluation results demonstrate SAMOSA's feasibility and efficacy in enhancing XR auditory realism.
Authors: C. C\'esar Claros Olivares, Austin J. Brockmeier
Abstract: We present a systematic comparison of out-of-distribution (OOD) detection methods across CLIP-stratified regimes using AURC and AUGRC as primary metrics. Experiments cover two representation paradigms: CNNs trained from scratch and a fine-tuned Vision Transformer (ViT), evaluated on CIFAR-10/100, SuperCIFAR-100, and TinyImageNet. Using a multiple-comparison-controlled, rank-based pipeline (Friedman test with Conover-Holm post-hoc) and Bron-Kerbosch cliques, we find that the learned feature space largely determines OOD efficacy. For both CNNs and ViTs, probabilistic scores (e.g., MSR, GEN) dominate misclassification (ID) detection. Under stronger shifts, geometry-aware scores (e.g., NNGuide, fDBD, CTM) prevail on CNNs, whereas on ViTs GradNorm and KPCA Reconstruction Error remain consistently competitive. We further show a class-count-dependent trade-off for Monte-Carlo Dropout (MCD) and that a simple PCA projection improves several detectors. These results support a representation-centric view of OOD detection and provide statistically grounded guidance for method selection under distribution shift.
Authors: Omar Abdelrazik, Mohamed Elsayed, Noorul Wahab, Nasir Rajpoot, Adam Shephard
Abstract: Ultrasound-based risk stratification of thyroid nodules is a critical clinical task, but it suffers from high inter-observer variability. While many deep learning (DL) models function as "black boxes," we propose a fully automated, two-stage framework for interpretable malignancy prediction. Our method achieves interpretability by forcing the model to focus only on clinically relevant regions. First, a TransUNet model automatically segments the thyroid nodule. The resulting mask is then used to create a region of interest around the nodule, and this localised image is fed directly into a ResNet-18 classifier. We evaluated our framework using 5-fold cross-validation on a clinical dataset of 349 images, where it achieved a high F1-score of 0.852 for predicting malignancy. To validate its performance, we compared it against a strong baseline using a Random Forest classifier with hand-crafted morphological features, which achieved an F1-score of 0.829. The superior performance of our DL framework suggests that the implicit visual features learned from the localised nodule are more predictive than explicit shape features alone. This is the first fully automated end-to-end pipeline for both detecting thyroid nodules on ultrasound images and predicting their malignancy.
Authors: Tenghao Ji, Eytan Adar
Abstract: A challenge in fine-tuning text-to-image diffusion models for specific topics is to select good examples. Fine-tuning from image sets of varying quality, such as Wikipedia Commons, will often produce poor output. However, training images that \textit{do} exemplify the target concept (e.g., a \textit{female Mountain Bluebird}) help ensure that the generated images are similarly representative (e.g., have the prototypical blue-wings and gray chest). In this work, we propose QZLoRA, a framework to select images for low-rank adaptation (LoRA). The approach leverages QuizRank, a method to automatically rank images by treating them as an `educational intervention' and `quizzing' a VLM. We demonstrate that QZLoRA can produce better aligned, photorealistic images with fewer samples. We also show that these fine-tuned models can produce stylized that are similarly representative (i.e., illustrations). Our results highlight the promise of combining automated visual reasoning with parameter-efficient fine-tuning for topic-adaptive generative modeling.
Authors: Yunqi Hong, Johnson Kao, Liam Edwards, Nein-Tzu Liu, Chung-Yen Huang, Alex Oliveira-Kowaleski, Cho-Jui Hsieh, Neil Y. C. Lin
Abstract: AI tools in pathology have improved screening throughput, standardized quantification, and revealed prognostic patterns that inform treatment. However, adoption remains limited because most systems still lack the human-readable reasoning needed to audit decisions and prevent errors. We present RECAP-PATH, an interpretable framework that establishes a self-learning paradigm, shifting off-the-shelf multimodal large language models from passive pattern recognition to evidence-linked diagnostic reasoning. At its core is a two-phase learning process that autonomously derives diagnostic criteria: diversification expands pathology-style explanations, while optimization refines them for accuracy. This self-learning approach requires only small labeled sets and no white-box access or weight updates to generate cancer diagnoses. Evaluated on breast and prostate datasets, RECAP-PATH produced rationales aligned with expert assessment and delivered substantial gains in diagnostic accuracy over baselines. By uniting visual understanding with reasoning, RECAP-PATH provides clinically trustworthy AI and demonstrates a generalizable path toward evidence-linked interpretation.
Authors: Wenxuan Miao, Yulin Sun, Aiyue Chen, Jing Lin, Yiwu Yao, Yiming Gan, Jieru Zhao, Jingwen Leng, Mingyi Guo, Yu Feng
Abstract: The recent surge in video generation has shown the growing demand for high-quality video synthesis using large vision models. Existing video generation models are predominantly based on the video diffusion transformer (vDiT), however, they suffer from substantial inference delay due to self-attention. While prior studies have focused on reducing redundant computations in self-attention, they often overlook the inherent spatio-temporal correlations in video streams and directly leverage sparsity patterns from large language models to reduce attention computations. In this work, we take a principled approach to accelerate self-attention in vDiTs by leveraging the spatio-temporal correlations in the latent space. We show that the attention patterns within vDiT are primarily due to the dominant spatial and temporal correlations at the token channel level. Based on this insight, we propose a lightweight and adaptive reuse strategy that approximates attention computations by reusing partial attention scores of spatially or temporally correlated tokens along individual channels. We demonstrate that our method achieves significantly higher computational savings (85\%) compared to state-of-the-art techniques over 4 vDiTs, while preserving almost identical video quality ($<$0.06\% loss on VBench).
Authors: Shanmin Wang, Dongdong Zhao
Abstract: Knowledge Distillation (KD) is essential for compressing large models, yet relying on pre-trained "teacher" models downloaded from third-party repositories introduces serious security risks -- most notably backdoor attacks. Existing KD backdoor methods are typically complex and computationally intensive: they employ surrogate student models and simulated distillation to guarantee transferability, and they construct triggers in a way similar to universal adversarial perturbations (UAPs), which being not stealthy in magnitude, inherently exhibit strong adversarial behavior. This work questions whether such complexity is necessary and constructs stealthy "weak" triggers -- imperceptible perturbations that have negligible adversarial effect. We propose BackWeak, a simple, surrogate-free attack paradigm. BackWeak shows that a powerful backdoor can be implanted by simply fine-tuning a benign teacher with a weak trigger using a very small learning rate. We demonstrate that this delicate fine-tuning is sufficient to embed a backdoor that reliably transfers to diverse student architectures during a victim's standard distillation process, yielding high attack success rates. Extensive empirical evaluations on multiple datasets, model architectures, and KD methods show that BackWeak is efficient, simpler, and often more stealthy than previous elaborate approaches. This work calls on researchers studying KD backdoor attacks to pay particular attention to the trigger's stealthiness and its potential adversarial characteristics.
Authors: Pinxue Guo, Chongruo Wu, Xinyu Zhou, Lingyi Hong, Zhaoyu Chen, Jinglun Li, Kaixun Jiang, Sen-ching Samson Cheung, Wei Zhang, Wenqiang Zhang
Abstract: Multimodal Large Language Models (MLLMs) have unlocked powerful cross-modal capabilities, but still significantly suffer from hallucinations. As such, accurate detection of hallucinations in MLLMs is imperative for ensuring their reliability in practical applications. To this end, guided by the principle of "Seeing is Believing", we introduce VBackChecker, a novel reference-free hallucination detection framework that verifies the consistency of MLLMgenerated responses with visual inputs, by leveraging a pixellevel Grounding LLM equipped with reasoning and referring segmentation capabilities. This reference-free framework not only effectively handles rich-context scenarios, but also offers interpretability. To facilitate this, an innovative pipeline is accordingly designed for generating instruction-tuning data (R-Instruct), featuring rich-context descriptions, grounding masks, and hard negative samples. We further establish R^2 -HalBench, a new hallucination benchmark for MLLMs, which, unlike previous benchmarks, encompasses real-world, rich-context descriptions from 18 MLLMs with high-quality annotations, spanning diverse object-, attribute, and relationship-level details. VBackChecker outperforms prior complex frameworks and achieves state-of-the-art performance on R^2 -HalBench, even rivaling GPT-4o's capabilities in hallucination detection. It also surpasses prior methods in the pixel-level grounding task, achieving over a 10% improvement. All codes, data, and models are available at https://github.com/PinxueGuo/VBackChecker.
Authors: Jialiang Wang, Xiong Zhou, Xianming Liu, Gangfeng Hu, Deming Zhai, Junjun Jiang, Haoliang Li
Abstract: Mitigating the negative impact of noisy labels has been aperennial issue in supervised learning. Robust loss functions have emerged as a prevalent solution to this problem. In this work, we introduce the Variation Ratio as a novel property related to the robustness of loss functions, and propose a new family of robust loss functions, termed Variation-Bounded Loss (VBL), which is characterized by a bounded variation ratio. We provide theoretical analyses of the variation ratio, proving that a smaller variation ratio would lead to better robustness. Furthermore, we reveal that the variation ratio provides a feasible method to relax the symmetric condition and offers a more concise path to achieve the asymmetric condition. Based on the variation ratio, we reformulate several commonly used loss functions into a variation-bounded form for practical applications. Positive experiments on various datasets exhibit the effectiveness and flexibility of our approach.
Authors: Jiayu Li, Yunhan Zhao, Xiang Zheng, Zonghuan Xu, Yige Li, Xingjun Ma, Yu-Gang Jiang
Abstract: Vision-Language-Action (VLA) models enable robots to interpret natural-language instructions and perform diverse tasks, yet their integration of perception, language, and control introduces new safety vulnerabilities. Despite growing interest in attacking such models, the effectiveness of existing techniques remains unclear due to the absence of a unified evaluation framework. One major issue is that differences in action tokenizers across VLA architectures hinder reproducibility and fair comparison. More importantly, most existing attacks have not been validated in real-world scenarios. To address these challenges, we propose AttackVLA, a unified framework that aligns with the VLA development lifecycle, covering data construction, model training, and inference. Within this framework, we implement a broad suite of attacks, including all existing attacks targeting VLAs and multiple adapted attacks originally developed for vision-language models, and evaluate them in both simulation and real-world settings. Our analysis of existing attacks reveals a critical gap: current methods tend to induce untargeted failures or static action states, leaving targeted attacks that drive VLAs to perform precise long-horizon action sequences largely unexplored. To fill this gap, we introduce BackdoorVLA, a targeted backdoor attack that compels a VLA to execute an attacker-specified long-horizon action sequence whenever a trigger is present. We evaluate BackdoorVLA in both simulated benchmarks and real-world robotic settings, achieving an average targeted success rate of 58.4% and reaching 100% on selected tasks. Our work provides a standardized framework for evaluating VLA vulnerabilities and demonstrates the potential for precise adversarial manipulation, motivating further research on securing VLA-based embodied systems.
Authors: Petr Boriskov, Kirill Rudkovskii, Andrei Velichko
Abstract: This paper studies the removal of salt-and-pepper noise from images using median filter (MF) and simple three-layer autoencoder (AE) within recursive threshold algorithm. The performance of denoising is assessed with two metrics: the standard Structural Similarity Index SSIMImg of restored and clean images and a newly applied metric SSIMMap - the SSIM of entropy maps of these images computed via 2D Sample Entropy in sliding windows. We shown that SSIMMap is more sensitive to blur and local intensity transitions and complements SSIMImg. Experiments on low- and high-resolution grayscales images demonstrate that recursive threshold MF robustly restores images even under strong noise (50-60 %), whereas simple AE is only capable of restoring images with low levels of noise (<30 %). We propose two scalable schemes: (i) 2MF, which uses two MFs with different window sizes and a final thresholding step, effective for highlighting sharp local details at low resolution; and (ii) MFs-AE, which aggregates features from multiple MFs via an AE and is beneficial for restoring the overall scene structure at higher resolution. Owing to its simplicity and computational efficiency, MF remains preferable for deployment on resource-constrained platforms (edge/IoT), whereas AE underperforms without prior denoising. The results also validate the practical value of SSIMMap for objective blur assessment and denoising parameter tuning.
Authors: Junhyuk Seo, Hyeyoon Moon, Kyu-Hwan Jung, Namkee Oh, Taerim Kim
Abstract: Unplanned extubation (UE) remains a critical patient safety concern in intensive care units (ICUs), often leading to severe complications or death. Real-time UE detection has been limited, largely due to the ethical and privacy challenges of obtaining annotated ICU video data. We propose Augmented Unplanned Removal Alert (AURA), a vision-based risk detection system developed and validated entirely on a fully synthetic video dataset. By leveraging text-to-video diffusion, we generated diverse and clinically realistic ICU scenarios capturing a range of patient behaviors and care contexts. The system applies pose estimation to identify two high-risk movement patterns: collision, defined as hand entry into spatial zones near airway tubes, and agitation, quantified by the velocity of tracked anatomical keypoints. Expert assessments confirmed the realism of the synthetic data, and performance evaluations showed high accuracy for collision detection and moderate performance for agitation recognition. This work demonstrates a novel pathway for developing privacy-preserving, reproducible patient safety monitoring systems with potential for deployment in intensive care settings.
Authors: Kerem Basim (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey), Mehmet Ozan Unal (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey), Metin Ertas (Istanbul University, Istanbul, Turkey), Isa Yildirim (Electronics and Communication Engineering Department, Istanbul Technical University, Istanbul, Turkey)
Abstract: Block-Matching and 3D Filtering (BM3D) exploits non-local self-similarity priors for denoising but relies on fixed parameters. Deep models such as U-Net are more flexible but often lack interpretability and fail to generalize across noise regimes. In this study, we propose Deep Unfolded BM3D (DU-BM3D), a hybrid framework that unrolls BM3D into a trainable architecture by replacing its fixed collaborative filtering with a learnable U-Net denoiser. This preserves BM3D's non-local structural prior while enabling end-to-end optimization. We evaluate DU-BM3D on low-dose CT (LDCT) denoising and show that it outperforms classic BM3D and standalone U-Net across simulated LDCT at different noise levels, yielding higher PSNR and SSIM, especially in high-noise conditions.
Authors: Elhadji Cisse Faye, Mame Diarra Fall, Nicolas Dobigeon, Eric Barat
Abstract: This paper proposes a novel Bayesian framework for solving Poisson inverse problems by devising a Monte Carlo sampling algorithm which accounts for the underlying non-Euclidean geometry. To address the challenges posed by the Poisson likelihood -- such as non-Lipschitz gradients and positivity constraints -- we derive a Bayesian model which leverages exact and asymptotically exact data augmentations. In particular, the augmented model incorporates two sets of splitting variables both derived through a Bregman divergence based on the Burg entropy. Interestingly the resulting augmented posterior distribution is characterized by conditional distributions which benefit from natural conjugacy properties and preserve the intrinsic geometry of the latent and splitting variables. This allows for efficient sampling via Gibbs steps, which can be performed explicitly for all conditionals, except the one incorporating the regularization potential. For this latter, we resort to a Hessian Riemannian Langevin Monte Carlo (HRLMC) algorithm which is well suited to handle priors with explicit or easily computable score functions. By operating on a mirror manifold, this Langevin step ensures that the sampling satisfies the positivity constraints and more accurately reflects the underlying problem structure. Performance results obtained on denoising, deblurring, and positron emission tomography (PET) experiments demonstrate that the method achieves competitive performance in terms of reconstruction quality compared to optimization- and sampling-based approaches.
Authors: Rui Wang, Zeming Wei, Xiyue Zhang, Meng Sun
Abstract: Deep Neural Networks (DNNs) are known to be vulnerable to various adversarial perturbations. To address the safety concerns arising from these vulnerabilities, adversarial training (AT) has emerged as one of the most effective paradigms for enhancing the robustness of DNNs. However, existing AT frameworks primarily focus on a single or a limited set of attack types, leaving DNNs still exposed to attack types that may be encountered in practice but not addressed during training. In this paper, we propose an efficient fine-tuning method called Calibrated Adversarial Sampling (CAS) to address these issues. From the optimization perspective within the multi-armed bandit framework, it dynamically designs rewards and balances exploration and exploitation by considering the dynamic and interdependent characteristics of multiple robustness dimensions. Experiments on benchmark datasets show that CAS achieves superior overall robustness while maintaining high clean accuracy, providing a new paradigm for robust generalization of DNNs.
Authors: Rupam Mukherjee, Rajkumar Daniel, Soujanya Hazra, Shirin Dasgupta, Subhamoy Mandal
Abstract: Early detection of oral cancer and potentially malignant disorders is challenging in low-resource settings due to limited annotated data. We present a unified four-class oral lesion classifier that integrates deep RGB embeddings, hyperspectral reconstruction, handcrafted spectral-textural descriptors, and demographic metadata. A pathologist-verified subset of oral cavity images was curated and processed using a fine-tuned ConvNeXt-v2 encoder, followed by RGB-to-HSI reconstruction into 31-band hyperspectral cubes. Haemoglobin-sensitive indices, texture features, and spectral-shape measures were extracted and fused with deep and clinical features. Multiple machine-learning models were assessed with patient-wise validation. We further introduce an incremental heuristic meta-learner (IHML) that combines calibrated base classifiers through probabilistic stacking and patient-level posterior smoothing. On an unseen patient split, the proposed framework achieved a macro F1 of 66.23% and an accuracy of 64.56%. Results demonstrate that hyperspectral reconstruction and uncertainty-aware meta-learning substantially improve robustness for real-world oral lesion screening.
Authors: Rupam Mukherjee, Rajkumar Daniel, Soujanya Hazra, Shirin Dasgupta, Subhamoy Mandal
Abstract: Cytology is a valuable tool for early detection of oral squamous cell carcinoma (OSCC). However, manual examination of cytology whole slide images (WSIs) is slow, subjective, and depends heavily on expert pathologists. To address this, we introduce the first weakly supervised deep learning framework for patient-level diagnosis of oral cytology whole slide images, leveraging the newly released Oral Cytology Dataset [1], which provides annotated cytology WSIs from ten medical centres across India. Each patient case is represented as a bag of cytology patches and assigned a diagnosis label (Healthy, Benign, Oral Potentially Malignant Disorders (OPMD), OSCC) by an in-house expert pathologist. These patient-level weak labels form a new extension to the dataset. We evaluate a baseline multiple-instance learning (MIL) model and a proposed Region-Affinity Attention MIL (RAA-MIL) that models spatial relationships between regions within each slide. The RAA-MIL achieves an average accuracy of 72.7%, weighted F1-score of 0.69 on an unseen test set, outperforming the baseline. This study establishes the first patient-level weakly supervised benchmark for oral cytology and moves toward reliable AI-assisted digital pathology.
Authors: Fan Li, Arun Iyengar, Lanyu Xu
Abstract: In the field of medical imaging, AI-assisted techniques such as object detection, segmentation, and classification are widely employed to alleviate the workload of physicians and doctors. However, single-task models are predominantly used, overlooking the shared information across tasks. This oversight leads to inefficiencies in real-life applications. In this work, we propose MTMed3D, a novel end-to-end Multi-task Transformer-based model to address the limitations of single-task models by jointly performing 3D detection, segmentation, and classification in medical imaging. Our model uses a Transformer as the shared encoder to generate multi-scale features, followed by CNN-based task-specific decoders. The proposed framework was evaluated on the BraTS 2018 and 2019 datasets, achieving promising results across all three tasks, especially in detection, where our method achieves better results than prior works. Additionally, we compare our multi-task model with equivalent single-task variants trained separately. Our multi-task model significantly reduces computational costs and achieves faster inference speed while maintaining comparable performance to the single-task models, highlighting its efficiency advantage. To the best of our knowledge, this is the first work to leverage Transformers for multi-task learning that simultaneously covers detection, segmentation, and classification tasks in 3D medical imaging, presenting its potential to enhance diagnostic processes. The code is available at https://github.com/fanlimua/MTMed3D.git.
Authors: Jiacheng Wang, Hao Li, Xing Yao, Ahmad Toubasi, Taegan Vinarsky, Caroline Gheen, Joy Derwenskus, Chaoyang Jin, Richard Dortch, Junzhong Xu, Francesca Bagnato, Ipek Oguz
Abstract: Quantitative magnetization transfer (qMT) imaging provides myelin-sensitive biomarkers, such as the pool size ratio (PSR), which is valuable for multiple sclerosis (MS) assessment. However, qMT requires specialized 20-30 minute scans. We propose DEMIST to synthesize PSR maps from standard T1w and FLAIR images using a 3D latent diffusion model with three complementary conditioning mechanisms. Our approach has two stages: first, we train separate autoencoders for PSR and anatomical images to learn aligned latent representations. Second, we train a conditional diffusion model in this latent space on top of a frozen diffusion foundation backbone. Conditioning is decoupled into: (i) \textbf{semantic} tokens via cross-attention, (ii) \textbf{spatial} per-scale residual hints via a 3D ControlNet branch, and (iii) \textbf{adaptive} LoRA-modulated attention. We include edge-aware loss terms to preserve lesion boundaries and alignment losses to maintain quantitative consistency, while keeping the number of trainable parameters low and retaining the inductive bias of the pretrained model. We evaluate on 163 scans from 99 subjects using 5-fold cross-validation. Our method outperforms VAE, GAN and diffusion baselines on multiple metrics, producing sharper boundaries and better quantitative agreement with ground truth. Our code is publicly available at https://github.com/MedICL-VU/MS-Synthesis-3DcLDM.
Authors: Yu Liang, Yu Yang, Wenjie Wei, Ammar Belatreche, Shuai Wang, Malu Zhang, Yang Yang
Abstract: Binary Spiking Neural Networks (BSNNs) offer promising efficiency advantages for resource-constrained computing. However, their training algorithms often require substantial memory overhead due to latent weights storage and temporal processing requirements. To address this issue, we propose Binary Spiking Online (BSO) optimization algorithm, a novel online training algorithm that significantly reduces training memory. BSO directly updates weights through flip signals under the online training framework. These signals are triggered when the product of gradient momentum and weights exceeds a threshold, eliminating the need for latent weights during training. To enhance performance, we propose T-BSO, a temporal-aware variant that leverages the inherent temporal dynamics of BSNNs by capturing gradient information across time steps for adaptive threshold adjustment. Theoretical analysis establishes convergence guarantees for both BSO and T-BSO, with formal regret bounds characterizing their convergence rates. Extensive experiments demonstrate that both BSO and T-BSO achieve superior optimization performance compared to existing training methods for BSNNs. The codes are available at https://github.com/hamings1/BSO.
Authors: David Denisov, Shlomi Dolev, Dan Felmdan, Michael Segal
Abstract: We study the $k$-means problem for a set $\mathcal{S} \subseteq \mathbb{R}^d$ of $n$ segments, aiming to find $k$ centers $X \subseteq \mathbb{R}^d$ that minimize $D(\mathcal{S},X) := \sum_{S \in \mathcal{S}} \min_{x \in X} D(S,x)$, where $D(S,x) := \int_{p \in S} |p - x| dp$ measures the total distance from each point along a segment to a center. Variants of this problem include handling outliers, employing alternative distance functions such as M-estimators, weighting distances to achieve balanced clustering, or enforcing unique cluster assignments. For any $\varepsilon > 0$, an $\varepsilon$-coreset is a weighted subset $C \subseteq \mathbb{R}^d$ that approximates $D(\mathcal{S},X)$ within a factor of $1 \pm \varepsilon$ for any set of $k$ centers, enabling efficient streaming, distributed, or parallel computation. We propose the first coreset construction that provably handles arbitrary input segments. For constant $k$ and $\varepsilon$, it produces a coreset of size $O(\log^2 n)$ computable in $O(nd)$ time. Experiments, including a real-time video tracking application, demonstrate substantial speedups with minimal loss in clustering accuracy, confirming both the practical efficiency and theoretical guarantees of our method.
Authors: Yunxin Li, Xinyu Chen, Shenyuan Jiang, Haoyuan Shi, Zhenyu Liu, Xuanyu Zhang, Nanhao Deng, Zhenran Xu, Yicheng Ma, Meishan Zhang, Baotian Hu, Min Zhang
Abstract: We present Uni-MoE 2.0 from the Lychee family. As a fully open-source omnimodal large model (OLM), it substantially advances Lychee's Uni-MoE series in language-centric multimodal understanding, reasoning, and generating. Based on the Qwen2.5-7B dense architecture, we build Uni-MoE-2.0-Omni from scratch through three core contributions: dynamic-capacity Mixture-of-Experts (MoE) design, a progressive training strategy enhanced with an iterative reinforcement strategy, and a carefully curated multimodal data matching technique. It is capable of omnimodal understanding, as well as generating images, text, and speech. Architecturally, our new MoE framework balances computational efficiency and capability for 10 cross-modal inputs using shared, routed, and null experts, while our Omni-Modality 3D RoPE ensures spatio-temporal cross-modality alignment in the self-attention layer. For training, following cross-modal pretraining, we use a progressive supervised fine-tuning strategy that activates modality-specific experts and is enhanced by balanced data composition and an iterative GSPO-DPO method to stabilise RL training and improve reasoning. Data-wise, the base model, trained on approximately 75B tokens of open-source multimodal data, is equipped with special speech and image generation tokens, allowing it to learn these generative tasks by conditioning its outputs on linguistic cues. Extensive evaluation across 85 benchmarks demonstrates that our model achieves SOTA or highly competitive performance against leading OLMs, surpassing Qwen2.5-Omni (trained with 1.2T tokens) on over 50 of 76 benchmarks. Key strengths include video understanding (+7% avg. of 8), omnimodallity understanding (+7% avg. of 4), and audiovisual reasoning (+4%). It also advances long-form speech processing (reducing WER by 4.2%) and leads in low-level image processing and controllable generation across 5 metrics.
Authors: Sushrut Thorat, Adrien Doerig, Alexander Kroner, Carmen Amme, Tim C. Kietzmann
Abstract: Scenes are complex, yet structured collections of parts, including objects and surfaces, that exhibit spatial and semantic relations to one another. An effective visual system therefore needs unified scene representations that relate scene parts to their location and their co-occurrence. We hypothesize that this structure can be learned self-supervised from natural experience by exploiting the temporal regularities of active vision: each fixation reveals a locally-detailed glimpse that is statistically related to the previous one via co-occurrence and saccade-conditioned spatial regularities. We instantiate this idea with Glimpse Prediction Networks (GPNs) -- recurrent models trained to predict the feature embedding of the next glimpse along human-like scanpaths over natural scenes. GPNs successfully learn co-occurrence structure and, when given relative saccade location vectors, show sensitivity to spatial arrangement. Furthermore, recurrent variants of GPNs were able to integrate information across glimpses into a unified scene representation. Notably, these scene representations align strongly with human fMRI responses during natural-scene viewing across mid/high-level visual cortex. Critically, GPNs outperform architecture- and dataset-matched controls trained with explicit semantic objectives, and match or exceed strong modern vision baselines, leaving little unique variance for those alternatives. These results establish next-glimpse prediction during active vision as a biologically plausible, self-supervised route to brain-aligned scene representations learned from natural visual experience.
Authors: Emilien Valat, Ozan \"Oktem
Abstract: Ensuring proper generalization is a critical challenge in applying data-driven methods for solving inverse problems in imaging, as neural networks reconstructing an image must perform well across varied datasets and acquisition geometries. In X-ray Computed Tomography (CT), convolutional neural networks (CNNs) are widely used to filter the projection data but are ill-suited for this task as they apply grid-based convolutions to the sinogram, which inherently lies on a line manifold, not a regular grid. The CNNs, unaware of the geometry, are implicitly tied to it and require an excessive amount of parameters as they must infer the relations between measurements from the data rather than from prior information. The contribution of this paper is twofold. First, we introduce a graph data structure to represent CT acquisition geometries and tomographic data, providing a detailed explanation of the graph's structure for circular, cone-beam geometries. Second, we propose GLM, a hybrid neural network architecture that leverages both graph and grid convolutions to process tomographic data. We demonstrate that GLM outperforms CNNs when performance is quantified in terms of structural similarity and peak signal-to-noise ratio, despite the fact that GLM uses only a fraction of the trainable parameters. Compared to CNNs, GLM also requires significantly less training time and memory, and its memory requirements scale better. Crucially, GLM demonstrates robust generalization to unseen variations in the acquisition geometry, like when training only on fully sampled CT data and then testing on sparse-view CT data.
Authors: Min Gu Kwak, Yeonju Lee, Hairong Wang, Jing Li
Abstract: Brain tumors are among the most clinically significant neurological diseases and remain a major cause of morbidity and mortality due to their aggressive growth and structural heterogeneity. As tumors expand, they induce substantial anatomical deformation that disrupts both local tissue organization and global brain architecture, complicating diagnosis, treatment planning, and surgical navigation. Yet a subject-specific reference of how the brain would appear without tumor-induced changes is fundamentally unobtainable in clinical practice. We present BrainNormalizer, an anatomy-informed diffusion framework that reconstructs pseudo-healthy MRIs directly from tumorous scans by conditioning the generative process on boundary cues extracted from the subject's own anatomy. This boundary-guided conditioning enables anatomically plausible pseudo-healthy reconstruction without requiring paired non-tumorous and tumorous scans. BrainNormalizer employs a two-stage training strategy. The pretrained diffusion model is first adapted through inpainting-based fine-tuning on tumorous and non-tumorous scans. Next, an edge-map-guided ControlNet branch is trained to inject fine-grained anatomical contours into the frozen decoder while preserving learned priors. During inference, a deliberate misalignment strategy pairs tumorous inputs with non-tumorous prompts and mirrored contralateral edge maps, leveraging hemispheric correspondence to guide reconstruction. On the BraTS2020 dataset, BrainNormalizer achieves strong quantitative performance and qualitatively produces anatomically plausible reconstructions in tumor-affected regions while retaining overall structural coherence. BrainNormalizer provides clinically reliable anatomical references for treatment planning and supports new research directions in counterfactual modeling and tumor-induced deformation analysis.
Authors: Wenxin Zhu, Andong Chen, Yuchen Song, Kehai Chen, Conghui Zhu, Ziyan Chen, Tiejun Zhao
Abstract: With the remarkable success of Multimodal Large Language Models (MLLMs) in perception tasks, enhancing their complex reasoning capabilities has emerged as a critical research focus. Existing models still suffer from challenges such as opaque reasoning paths and insufficient generalization ability. Chain-of-Thought (CoT) reasoning, which has demonstrated significant efficacy in language models by enhancing reasoning transparency and output interpretability, holds promise for improving model reasoning capabilities when extended to the multimodal domain. This paper provides a systematic review centered on "Multimodal Chain-of-Thought" (MCoT). First, it analyzes the background and theoretical motivations for its inception from the perspectives of technical evolution and task demands. Then, it introduces mainstream MCoT methods from three aspects: CoT paradigms, the post-training stage, and the inference stage, while also analyzing their underlying mechanisms. Furthermore, the paper summarizes existing evaluation benchmarks and metrics, and discusses the application scenarios of MCoT. Finally, it analyzes the challenges currently facing MCoT and provides an outlook on its future research directions.
Authors: Zhiqi Li, Yuchen Sun, Greg Turk, Bo Zhu
Abstract: We present Functional Mean Flow (FMF) as a one-step generative model defined in infinite-dimensional Hilbert space. FMF extends the one-step Mean Flow framework to functional domains by providing a theoretical formulation for Functional Flow Matching and a practical implementation for efficient training and sampling. We also introduce an $x_1$-prediction variant that improves stability over the original $u$-prediction form. The resulting framework is a practical one-step Flow Matching method applicable to a wide range of functional data generation tasks such as time series, images, PDEs, and 3D geometry.
Authors: Changhun Oh, Seongryong Oh, Jinwoo Hwang, Yoonsung Kim, Hardik Sharma, Jongse Park
Abstract: 3D Gaussian Splatting (3DGS) rendering in real-time on resource-constrained devices is essential for delivering immersive augmented and virtual reality (AR/VR) experiences. However, existing solutions struggle to achieve high frame rates, especially for high-resolution rendering. Our analysis identifies the sorting stage in the 3DGS rendering pipeline as the major bottleneck due to its high memory bandwidth demand. This paper presents Neo, which introduces a reuse-and-update sorting algorithm that exploits temporal redundancy in Gaussian ordering across consecutive frames, and devises a hardware accelerator optimized for this algorithm. By efficiently tracking and updating Gaussian depth ordering instead of re-sorting from scratch, Neo significantly reduces redundant computations and memory bandwidth pressure. Experimental results show that Neo achieves up to 10.0x and 5.6x higher throughput than state-of-the-art edge GPU and ASIC solution, respectively, while reducing DRAM traffic by 94.5% and 81.3%. These improvements make high-quality and low-latency on-device 3D rendering more practical.
Authors: Guoyan Wang, Yanyan Huang, Chunlin Chen, Lifeng Wang, Yuxiang Sun
Abstract: Automated operation in cross-platform strategy games demands agents with robust generalization across diverse user interfaces and dynamic battlefield conditions. While vision-language models (VLMs) have shown considerable promise in multimodal reasoning, their application to complex human-computer interaction scenarios--such as strategy gaming--remains largely unexplored. Here, we introduce Yanyun-3, a general-purpose agent framework that, for the first time, enables autonomous cross-platform operation across three heterogeneous strategy game environments. By integrating the vision-language reasoning of Qwen2.5-VL with the precise execution capabilities of UI-TARS, Yanyun-3 successfully performs core tasks including target localization, combat resource allocation, and area control. Through systematic ablation studies, we evaluate the effects of various multimodal data combinations--static images, multi-image sequences, and videos--and propose the concept of combination granularity to differentiate between intra-sample fusion and inter-sample mixing strategies. We find that a hybrid strategy, which fuses multi-image and video data while mixing in static images (MV+S), substantially outperforms full fusion: it reduces inference time by 63% and boosts the BLEU-4 score by a factor of 12 (from 4.81% to 62.41%, approximately 12.98x). Operating via a closed-loop pipeline of screen capture, model inference, and action execution, the agent demonstrates strong real-time performance and cross-platform generalization. Beyond providing an efficient solution for strategy game automation, our work establishes a general paradigm for enhancing VLM performance through structured multimodal data organization, offering new insights into the interplay between static perception and dynamic reasoning in embodied intelligence.
Authors: Pritam P. Karmokar, William J. Beksi
Abstract: Event cameras, by virtue of their working principle, directly encode motion within a scene. Many learning-based and model-based methods exist that estimate event-based optical flow, however the temporally dense yet spatially sparse nature of events poses significant challenges. To address these issues, contrast maximization (CM) is a prominent model-based optimization methodology that estimates the motion trajectories of events within an event volume by optimally warping them. Since its introduction, the CM framework has undergone a series of refinements by the computer vision community. Nonetheless, it remains a highly non-convex optimization problem. In this paper, we introduce a novel biologically-inspired hybrid CM method for event-based optical flow estimation that couples visual and inertial motion cues. Concretely, we propose the use of orientation maps, derived from camera 3D velocities, as priors to guide the CM process. The orientation maps provide directional guidance and constrain the space of estimated motion trajectories. We show that this orientation-guided formulation leads to improved robustness and convergence in event-based optical flow estimation. The evaluation of our approach on the MVSEC, DSEC, and ECD datasets yields superior accuracy scores over the state of the art.
Authors: Xuankun Rong, Wenke Huang, Tingfeng Wang, Daiguo Zhou, Bo Du, Mang Ye
Abstract: Multimodal large language models (MLLMs) have demonstrated impressive reasoning and instruction-following capabilities, yet their expanded modality space introduces new compositional safety risks that emerge from complex text-image interactions. Such cross-modal couplings can produce unsafe semantics even when individual inputs are benign, exposing the fragile safety awareness of current MLLMs. While recent works enhance safety by guiding models to reason about potential risks, unregulated reasoning traces may compromise alignment; although Group Relative Policy Optimization (GRPO) offers self-rewarded refinement without human supervision, it lacks verifiable signals for reasoning safety. To address this, we propose SafeGRPO a self-rewarded multimodal safety alignment framework that integrates rule-governed reward construction into GRPO, enabling interpretable and verifiable optimization of reasoning safety. Built upon the constructed SafeTag-VL-3K dataset with explicit visual, textual, and combined safety tags, SafeGRPO performs step-guided safety thinking to enforce structured reasoning and behavior alignment, substantially improving multimodal safety awareness, compositional robustness, and reasoning stability across diverse benchmarks without sacrificing general capabilities.
Authors: Minsoo Jo, Dongyoon Yang, Taesup Kim
Abstract: Adversarial examples in neural networks have been extensively studied in Euclidean geometry, but recent advances in \textit{hyperbolic networks} call for a reevaluation of attack strategies in non-Euclidean geometries. Existing methods such as FGSM and PGD apply perturbations without regard to the underlying hyperbolic structure, potentially leading to inefficient or geometrically inconsistent attacks. In this work, we propose a novel adversarial attack that explicitly leverages the geometric properties of hyperbolic space. Specifically, we compute the gradient of the loss function in the tangent space of hyperbolic space, decompose it into a radial (depth) component and an angular (semantic) component, and apply perturbation derived solely from the angular direction. Our method generates adversarial examples by focusing perturbations in semantically sensitive directions encoded in angular movement within the hyperbolic geometry. Empirical results on image classification, cross-modal retrieval tasks and network architectures demonstrate that our attack achieves higher fooling rates than conventional adversarial attacks, while producing high-impact perturbations with deeper insights into vulnerabilities of hyperbolic embeddings. This work highlights the importance of geometry-aware adversarial strategies in curved representation spaces and provides a principled framework for attacking hierarchical embeddings.
Authors: Harrison H. Li, Medhanie Irgau, Nabil Janmohamed, Karen Solveig Rieckmann, David B. Lobell
Abstract: Precise estimation and uncertainty quantification for average crop yields are critical for agricultural monitoring and decision making. Existing data collection methods, such as crop cuts in randomly sampled fields at harvest time, are relatively time-consuming. Thus, we propose an approach based on prediction-powered inference (PPI) to supplement these crop cuts with less time-consuming field photos. After training a computer vision model to predict the ground truth crop cut yields from the photos, we learn a ``control function" that recalibrates these predictions with the spatial coordinates of each field. This enables fields with photos but not crop cuts to be leveraged to improve the precision of zone-wide average yield estimates. Our control function is learned by training on a dataset of nearly 20,000 real crop cuts and photos of rice and maize fields in sub-Saharan Africa. To improve precision, we pool training observations across different zones within the same first-level subdivision of each country. Our final PPI-based point estimates of the average yield are provably asymptotically unbiased and cannot increase the asymptotic variance beyond that of the natural baseline estimator -- the sample average of the crop cuts -- as the number of fields grows. We also propose a novel bias-corrected and accelerated (BCa) bootstrap to construct accompanying confidence intervals. Even in zones with as few as 20 fields, the point estimates show significant empirical improvement over the baseline, increasing the effective sample size by as much as 73% for rice and by 12-23% for maize. The confidence intervals are accordingly shorter at minimal cost to empirical finite-sample coverage. This demonstrates the potential for relatively low-cost images to make area-based crop insurance more affordable and thus spur investment into sustainable agricultural practices.
Authors: Yong Liu, Keyang Ye, Tianjia Shao, Kun Zhou
Abstract: We propose Transmission-Reflection Gaussians (TR-Gaussians), a novel 3D-Gaussian-based representation for high-fidelity rendering of planar transmission and reflection, which are ubiquitous in indoor scenes. Our method combines 3D Gaussians with learnable reflection planes that explicitly model the glass planes with view-dependent reflectance strengths. Real scenes and transmission components are modeled by 3D Gaussians and the reflection components are modeled by the mirrored Gaussians with respect to the reflection plane. The transmission and reflection components are blended according to a Fresnel-based, view-dependent weighting scheme, allowing for faithful synthesis of complex appearance effects under varying viewpoints. To effectively optimize TR-Gaussians, we develop a multi-stage optimization framework incorporating color and geometry constraints and an opacity perturbation mechanism. Experiments on different datasets demonstrate that TR-Gaussians achieve real-time, high-fidelity novel view synthesis in scenes with planar transmission and reflection, and outperform state-of-the-art approaches both quantitatively and qualitatively.
Authors: Kyunghyun Lee, Yong-Min Shin, Minwoo Shin, Jihun Kim, Sunghwan Lim, Won-Yong Shin, Kyungho Yoon
Abstract: Early diagnosis of breast cancer is crucial, enabling the establishment of appropriate treatment plans and markedly enhancing patient prognosis. While direct magnetic resonance imaging-guided biopsy demonstrates promising performance in detecting cancer lesions, its practical application is limited by prolonged procedure times and high costs. To overcome these issues, an indirect MRI-guided biopsy that allows the procedure to be performed outside of the MRI room has been proposed, but it still faces challenges in creating an accurate real-time deformable breast model. In our study, we tackled this issue by developing a graph neural network (GNN)-based model capable of accurately predicting deformed breast cancer sites in real time during biopsy procedures. An individual-specific finite element (FE) model was developed by incorporating magnetic resonance (MR) image-derived structural information of the breast and tumor to simulate deformation behaviors. A GNN model was then employed, designed to process surface displacement and distance-based graph data, enabling accurate prediction of overall tissue displacement, including the deformation of the tumor region. The model was validated using phantom and real patient datasets, achieving an accuracy within 0.2 millimeters (mm) for cancer node displacement (RMSE) and a dice similarity coefficient (DSC) of 0.977 for spatial overlap with actual cancerous regions. Additionally, the model enabled real-time inference and achieved a speed-up of over 4,000 times in computational cost compared to conventional FE simulations. The proposed deformation-aware GNN model offers a promising solution for real-time tumor displacement prediction in breast biopsy, with high accuracy and real-time capability. Its integration with clinical procedures could significantly enhance the precision and efficiency of breast cancer diagnosis.
Authors: SeokJoo Kwak, Jihoon Kim, Boyoun Kim, Jung Jae Yoon, Wooseok Jang, Jeonghoon Hong, Jaeho Yang, Yeong-Dae Kwon
Abstract: Graphical User Interface (GUI) grounding - the task of mapping natural language instructions to screen coordinates - is essential for autonomous agents and accessibility technologies. Existing systems rely on monolithic models or one-shot pipelines that lack modularity and fail under visual clutter and ambiguous instructions. We introduce MEGA-GUI, a multi-stage framework that separates grounding into coarse Region-of-Interest (ROI) selection and fine-grained element grounding, orchestrated by specialized vision-language agents. MEGA-GUI features a bidirectional ROI zoom algorithm that mitigates spatial dilution and a context-aware rewriting agent that reduces semantic ambiguity. Our analysis reveals complementary strengths and weaknesses across vision-language models at different visual scales, and we show that leveraging this modular structure achieves consistently higher accuracy than monolithic approaches. On the visually dense ScreenSpot-Pro benchmark, MEGA-GUI attains 73.18% accuracy, and on the semantically complex OSWorld-G benchmark it reaches 68.63%, surpassing previously reported results. Code and the Grounding Benchmark Toolkit (GBT) are available at https://github.com/samsungsds-research-papers/mega-gui.
URLs: https://github.com/samsungsds-research-papers/mega-gui.
Authors: Gagan Raj Gupta, Anshul Kumar, Manish Rai, Apu Chakraborty, Ashutosh Modi, Abdelaali Chaoub, Soumajit Pramanik, Moyank Giri, Yashwanth Holla, Sunny Kumar, M. V. Kiran Sooraj
Abstract: Large Language Models (LLMs) have emerged as powerful tools for automating complex reasoning and decision-making tasks. In telecommunications, they hold the potential to transform network optimization, automate troubleshooting, enhance customer support, and ensure regulatory compliance. However, their deployment in telecom is hindered by domain-specific challenges that demand specialized adaptation. To overcome these challenges and to accelerate the adaptation of LLMs for telecom, we propose MM-Telco, a comprehensive suite of multimodal benchmarks and models tailored for the telecom domain. The benchmark introduces various tasks (both text based and image based) that address various practical real-life use cases such as network operations, network management, improving documentation quality, and retrieval of relevant text and images. Further, we perform baseline experiments with various LLMs and VLMs. The models fine-tuned on our dataset exhibit a significant boost in performance. Our experiments also help analyze the weak areas in the working of current state-of-art multimodal LLMs, thus guiding towards further development and research.
Authors: Cheng Peng, Zhenzhe Zhang, Cheng Chi, Xiaobao Wei, Yanhao Zhang, Heng Wang, Pengwei Wang, Zhongyuan Wang, Jing Liu, Shanghang Zhang
Abstract: Navigating to a specified object in an unknown environment is a fundamental yet challenging capability of embodied intelligence. However, current methods struggle to balance decision frequency with intelligence, resulting in decisions lacking foresight or discontinuous actions. In this work, we propose PIGEON: Point of Interest Guided Exploration for Object Navigation with VLM, maintaining a lightweight and semantically aligned snapshot memory during exploration as semantic input for the exploration strategy. We use a large Visual-Language Model (VLM), named PIGEON-VL, to select Points of Interest (PoI) formed during exploration and then employ a lower-level planner for action output, increasing the decision frequency. Additionally, this PoI-based decision-making enables the generation of Reinforcement Learning with Verifiable Reward (RLVR) data suitable for simulators. Experiments on classic object navigation benchmarks demonstrate that our zero-shot transfer method achieves state-of-the-art performance, while RLVR further enhances the model's semantic guidance capabilities, enabling deep reasoning during real-time navigation.
Authors: Xiaoqi Han, Ru Li, Ran Yi, Hongye Tan, Zhuomin Liang, V\'ictor Guti\'errez-Basulto, Jeff Z. Pan
Abstract: Multimodal Model Editing (MMED) aims to correct erroneous knowledge in multimodal models. Existing evaluation methods, adapted from textual model editing, overstate success by relying on low-similarity or random inputs, obscure overfitting. We propose a comprehensive locality evaluation framework, covering three key dimensions: random-image locality, no-image locality, and consistent-image locality, operationalized through seven distinct data types, enabling a detailed and structured analysis of multimodal edits. We introduce De-VQA, a dynamic evaluation for visual question answering, uncovering a phenomenon we term transient blindness, overfitting to edit-similar text while ignoring visuals. Token analysis shows edits disproportionately affect textual tokens. We propose locality-aware adversarial losses to balance cross-modal representations. Empirical results demonstrate that our approach consistently outperforms existing baselines, reducing transient blindness and improving locality by 17% on average.
Authors: Bowen Ye, Bin Zhang, Hang Zhao
Abstract: Gaining sustainable performance improvement with scaling data and model budget remains a pivotal yet unresolved challenge in autonomous driving. While autoregressive models exhibited promising data-scaling efficiency in planning tasks, predicting ego trajectories alone suffers sparse supervision and weakly constrains how scene evolution should shape ego motion. Therefore, we introduce DAP, a discrete-token autoregressive planner that jointly forecasts BEV semantics and ego trajectories, thereby enforcing comprehensive representation learning and allowing predicted dynamics to directly condition ego motion. In addition, we incorporate a reinforcement-learning-based fine-tuning, which preserves supervised behavior cloning priors while injecting reward-guided improvements. Despite a compact 160M parameter budget, DAP achieves state-of-the-art performance on open-loop metrics and delivers competitive closed-loop results on the NAVSIM benchmark. Overall, the fully discrete-token autoregressive formulation operating on both rasterized BEV and ego actions provides a compact yet scalable planning paradigm for autonomous driving.
Authors: Wanqing Cui, Wei Huang, Yazhi Guo, Yibo Hu, Meiguang Jin, Junfeng Ma, Keping Bi
Abstract: Visual document retrieval requires understanding heterogeneous and multi-modal content to satisfy information needs. Recent advances use screenshot-based document encoding with fine-grained late interaction, significantly improving retrieval performance. However, retrievers are still trained with coarse global relevance labels, without revealing which regions support the match. As a result, retrievers tend to rely on surface-level cues and struggle to capture implicit semantic connections, hindering their ability to handle non-extractive queries. To alleviate this problem, we propose a \textbf{A}ttention-\textbf{G}rounded \textbf{RE}triever \textbf{E}nhancement (AGREE) framework. AGREE leverages cross-modal attention from multimodal large language models as proxy local supervision to guide the identification of relevant document regions. During training, AGREE combines local signals with the global signals to jointly optimize the retriever, enabling it to learn not only whether documents match, but also which content drives relevance. Experiments on the challenging ViDoRe V2 benchmark show that AGREE significantly outperforms the global-supervision-only baseline. Quantitative and qualitative analyses further demonstrate that AGREE promotes deeper alignment between query terms and document regions, moving beyond surface-level matching toward more accurate and interpretable retrieval. Our code is available at: https://anonymous.4open.science/r/AGREE-2025.
Authors: Agnese Chiatti, Lara Piccolo, Sara Bernardini, Matteo Matteucci, Viola Schiaffonati
Abstract: With the growing deployment of Vision-Language Models (VLMs), pre-trained on large image-text and video-text datasets, it is critical to equip users with the tools to discern when to trust these systems. However, examining how user trust in VLMs builds and evolves remains an open problem. This problem is exacerbated by the increasing reliance on AI models as judges for experimental validation, to bypass the cost and implications of running participatory design studies directly with users. Following a user-centred approach, this paper presents preliminary results from a workshop with prospective VLM users. Insights from this pilot workshop inform future studies aimed at contextualising trust metrics and strategies for participants' engagement to fit the case of user-VLM interaction.
Authors: Pascal Zimmer, Ghassan Karame
Abstract: In this paper, we present the first detailed analysis of how optimization hyperparameters -- such as learning rate, weight decay, momentum, and batch size -- influence robustness against both transfer-based and query-based attacks. Supported by theory and experiments, our study spans a variety of practical deployment settings, including centralized training, ensemble learning, and distributed training. We uncover a striking dichotomy: for transfer-based attacks, decreasing the learning rate significantly enhances robustness by up to $64\%$. In contrast, for query-based attacks, increasing the learning rate consistently leads to improved robustness by up to $28\%$ across various settings and data distributions. Leveraging these findings, we explore -- for the first time -- the optimization hyperparameter design space to jointly enhance robustness against both transfer-based and query-based attacks. Our results reveal that distributed models benefit the most from hyperparameter tuning, achieving a remarkable tradeoff by simultaneously mitigating both attack types more effectively than other training setups.
Authors: Hyunwoo Oh, Hanning Chen, Sanggeon Yun, Yang Ni, Wenjun Huang, Tamoghno Das, Suyeon Jang, Mohsen Imani
Abstract: Deformable transformers deliver state-of-the-art detection but map poorly to hardware due to irregular memory access and low arithmetic intensity. We introduce QUILL, a schedule-aware accelerator that turns deformable attention into cache-friendly, single-pass work. At its core, Distance-based Out-of-Order Querying (DOOQ) orders queries by spatial proximity; the look-ahead drives a region prefetch into an alternate buffer--forming a schedule-aware prefetch loop that overlaps memory and compute. A fused MSDeformAttn engine executes interpolation, Softmax, aggregation, and the final projection (W''m) in one pass without spilling intermediates, while small tensors are kept on-chip and surrounding dense layers run on integrated GEMMs. Implemented as RTL and evaluated end-to-end, QUILL achieves up to 7.29x higher throughput and 47.3x better energy efficiency than an RTX 4090, and exceeds prior accelerators by 3.26-9.82x in throughput and 2.01-6.07x in energy efficiency. With mixed-precision quantization, accuracy tracks FP32 within <=0.9 AP across Deformable and Sparse DETR variants. By converting sparsity into locality--and locality into utilization--QUILL delivers consistent, end-to-end speedups.
Authors: Sofia Jamil, Kotla Sai Charan, Sriparna Saha, Koustava Goswami, Joseph K J
Abstract: Indian poetry, known for its linguistic complexity and deep cultural resonance, has a rich and varied heritage spanning thousands of years. However, its layered meanings, cultural allusions, and sophisticated grammatical constructions often pose challenges for comprehension, especially for non-native speakers or readers unfamiliar with its context and language. Despite its cultural significance, existing works on poetry have largely overlooked Indian language poems. In this paper, we propose the Translation and Image Generation (TAI) framework, leveraging Large Language Models (LLMs) and Latent Diffusion Models through appropriate prompt tuning. Our framework supports the United Nations Sustainable Development Goals of Quality Education (SDG 4) and Reduced Inequalities (SDG 10) by enhancing the accessibility of culturally rich Indian-language poetry to a global audience. It includes (1) a translation module that uses an Odds Ratio Preference Alignment Algorithm to accurately translate morphologically rich poetry into English, and (2) an image generation module that employs a semantic graph to capture tokens, dependencies, and semantic relationships between metaphors and their meanings, to create visually meaningful representations of Indian poems. Our comprehensive experimental evaluation, including both human and quantitative assessments, demonstrates the superiority of TAI Diffusion in poem image generation tasks, outperforming strong baselines. To further address the scarcity of resources for Indian-language poetry, we introduce the Morphologically Rich Indian Language Poems MorphoVerse Dataset, comprising 1,570 poems across 21 low-resource Indian languages. By addressing the gap in poetry translation and visual comprehension, this work aims to broaden accessibility and enrich the reader's experience.
Authors: Mahdi Pourmirzaei, Gholam Ali Montazer, Farzaneh Esmaili
Abstract: Facial emotion recognition (FER) is a fine-grained problem where the value of transfer learning is often assumed. We first quantify this assumption and show that, on AffectNet, training from random initialization with sufficiently strong augmentation consistently matches or surpasses fine-tuning from ImageNet. Motivated by this result, we propose Hybrid Multi-Task Learning (HMTL) for FER in the wild. HMTL augments supervised learning (SL) with self-supervised learning (SSL) objectives during training, while keeping the inference-time model unchanged. We instantiate HMTL with two tailored pretext tasks, puzzling and inpainting with a perceptual loss, that encourage part-aware and expression-relevant features. On AffectNet, both HMTL variants achieve state-of-the-art accuracy in the eight-emotion setting without any additional pretraining data, and they provide larger gains under low-data regimes. Compared with conventional SSL pretraining, HMTL yields stronger downstream performance. Beyond FER, the same strategy improves fine-grained facial analysis tasks, including head pose estimation and gender recognition. These results suggest that aligned SSL auxiliaries are an effective and simple way to strengthen supervised fine-grained facial representation without adding extra computation cost during inference time.
Authors: Zhongying Deng (and for the Alzheimer's Disease Neuroimaging Initiative), Shujun Wang (and for the Alzheimer's Disease Neuroimaging Initiative), Angelica I Aviles-Rivero (and for the Alzheimer's Disease Neuroimaging Initiative), Zoe Kourtzi (and for the Alzheimer's Disease Neuroimaging Initiative), Carola-Bibiane Sch\"onlieb (and for the Alzheimer's Disease Neuroimaging Initiative)
Abstract: Alzheimer's disease progression prediction is critical for patients with early Mild Cognitive Impairment (MCI) to enable timely intervention and improve their quality of life. While existing progression prediction techniques demonstrate potential with multimodal data, they are highly limited by their reliance on labelled data and fail to account for a key element of future progression prediction: not all features extracted at the current moment may be relevant for predicting progression several years later. To address these limitations in the literature, we design a novel semi-supervised multimodal learning hypergraph architecture, termed HIBMatch, by harnessing hypergraph knowledge based on information bottleneck and consistency regularisation strategies. Firstly, our framework utilises hypergraphs to represent multimodal data, encompassing both imaging and non-imaging modalities. Secondly, to harmonise relevant information from the currently captured data for future MCI conversion prediction, we propose a Hypergraph Information Bottleneck (HIB) that discriminates against irrelevant information, thereby focusing exclusively on harmonising relevant information for future MCI conversion prediction. Thirdly, our method enforces consistency regularisation between the HIB and a discriminative classifier to enhance the robustness and generalisation capabilities of HIBMatch under both topological and feature perturbations. Finally, to fully exploit the unlabeled data, HIBMatch incorporates a cross-modal contrastive loss for data efficiency. Extensive experiments on the Alzheimer's Disease Neuroimaging Initiative (ADNI) dataset demonstrate that our proposed HIBMatch framework surpasses existing state-of-the-art methods in Alzheimer's disease prognosis.
Authors: Jiang Liu, Chun Pong Lau, Zhongliang Guo, Yuxiang Guo, Zhaoyang Wang, Rama Chellappa
Abstract: The increasingly pervasive facial recognition (FR) systems raise serious concerns about personal privacy, especially for billions of users who have publicly shared their photos on social media. Several attempts have been made to protect individuals from being identified by unauthorized FR systems utilizing adversarial attacks to generate encrypted face images. However, existing methods suffer from poor visual quality or low attack success rates, which limit their utility. Recently, diffusion models have achieved tremendous success in image generation. In this work, we ask: can diffusion models be used to generate adversarial examples to improve both visual quality and attack performance? We propose DiffProtect, which utilizes a diffusion autoencoder to generate semantically meaningful perturbations on FR systems. Extensive experiments demonstrate that DiffProtect produces more natural-looking encrypted images than state-of-the-art methods while achieving significantly higher attack success rates, e.g., 24.5% and 25.1% absolute improvements on the CelebA-HQ and FFHQ datasets.
Authors: Stefano Woerner, Arthur Jaques, Christian F. Baumgartner
Abstract: While the field of medical image analysis has undergone a transformative shift with the integration of machine learning techniques, the main challenge of these techniques is often the scarcity of large, diverse, and well-annotated datasets. Medical images vary in format, size, and other parameters and therefore require extensive preprocessing and standardization, for usage in machine learning. Addressing these challenges, we introduce the Medical Imaging Meta-Dataset (MedIMeta), a novel multi-domain, multi-task meta-dataset. MedIMeta contains 19 medical imaging datasets spanning 10 different domains and encompassing 54 distinct medical tasks, all of which are standardized to the same format and readily usable in PyTorch or other ML frameworks. We perform a technical validation of MedIMeta, demonstrating its utility through fully supervised and cross-domain few-shot learning baselines.
Authors: Antonio Ruiz, Andrew Melnik, Nicolo Savioli, Dong Wang, Yanfeng Zhang, Helge Ritter
Abstract: The lane graph is critical for applications such as autonomous driving and lane-level route planning. While previous research has focused on extracting lane-level graphs from aerial imagery using convolutional neural networks (CNNs) followed by post-processing segmentation-to-graph algorithms, these methods often face challenges in producing sharp and complete segmentation masks. Challenges such as occlusions, variations in lighting, and changes in road texture can lead to incomplete and inaccurate lane masks, resulting in poor-quality lane graphs. To address these challenges, we propose a novel approach that refines the lane masks, output by a CNN, using diffusion models. Experimental results on a publicly available dataset demonstrate that our method outperforms existing methods based solely on CNNs or diffusion models, particularly in terms of graph connectivity. Our lane mask refinement approach enhances the quality of the extracted lane graph, yielding gains of approximately 1.5\% in GEO F1 and 3.5\% in TOPO F1 scores over the best-performing CNN-based method, and improvements of 28\% and 34\%, respectively, compared to a prior diffusion-based approach. Both GEO F1 and TOPO F1 scores are critical metrics for evaluating lane graph quality. Additionally, ablation studies are conducted to evaluate the individual components of our approach, providing insights into their respective contributions and effectiveness.
Authors: Taewon Kang, Divya Kothandaraman, Dinesh Manocha, Ming C. Lin
Abstract: Recent 3D novel view synthesis (NVS) methods often require extensive 3D data for training, and also typically lack generalization beyond the training distribution. Moreover, they tend to be object centric and struggle with complex and intricate scenes. Conversely, 3D-free methods can generate text-controlled views of complex, in-the-wild scenes using a pretrained stable diffusion model without the need for a large amount of 3D-based training data, but lack camera control. In this paper, we introduce a method capable of generating camera-controlled viewpoints from a single input image, by combining the benefits of 3D-free and 3D-based approaches. Our method excels in handling complex and diverse scenes without extensive training or additional 3D and multiview data. It leverages widely available pretrained NVS models for weak guidance, integrating this knowledge into a 3D-free view synthesis style approach, along with enriching the CLIP vision-language space with 3D camera angle information, to achieve the desired results. Experimental results demonstrate that our method outperforms existing models in both qualitative and quantitative evaluations, achieving high-fidelity, consistent novel view synthesis at desired camera angles across a wide variety of scenes while maintaining accurate, natural detail representation and image clarity across various viewpoints. We also support our method with a comprehensive analysis of 2D image generation models and the 3D space, providing a solid foundation and rationale for our solution.
Authors: Cheng-Yi Lee, Yu-Hsuan Chiang, Zhong-You Wu, Chia-Mu Yu, Chun-Shien Lu
Abstract: Visual State Space Models (VSSM) have shown remarkable performance in various computer vision tasks. However, backdoor attacks pose significant security challenges, causing compromised models to predict target labels when specific triggers are present while maintaining normal behavior on benign samples. In this paper, we investigate the robustness of VSSMs against backdoor attacks. Specifically, we delicately design a novel framework for VSSMs, dubbed BadVim, which utilizes low-rank perturbations on state-wise to uncover their impact on state transitions during training. By poisoning only $0.3\%$ of the training data, our attacks cause any trigger-embedded input to be misclassified to the targeted class with a high attack success rate (over 97%) at inference time. Our findings suggest that the state-space representation property of VSSMs, which enhances model capability, may also contribute to its vulnerability to backdoor attacks. Our attack exhibits effectiveness across three datasets, even bypassing state-of-the-art defenses against such attacks. Extensive experiments show that the backdoor robustness of VSSMs is comparable to that of Transformers (ViTs) and superior to that of Convolutional Neural Networks (CNNs). We believe our findings will prompt the community to reconsider the trade-offs between performance and robustness in model design.
Authors: Gustav Grund Pihlgren, Kary Fr\"amling
Abstract: Perturbation-based post-hoc image explanation methods are commonly used to explain image prediction models. These methods perturb parts of the input to measure how those parts affect the output. Since the methods only require the input and output, they can be applied to any model, making them a popular choice to explain black-box models. While many different methods exist and have been compared with one another, it remains poorly understood which parameters of the different methods are responsible for their varying performance. This work uses the Randomized Input Sampling for Explanations (RISE) method as a baseline to evaluate many combinations of mask sampling, segmentation techniques, smoothing, attribution calculation, and per-segment or per-pixel attribution, using a proxy metric. The results show that attribution calculation, which is frequently the focus of other works, has little impact on the results. Conversely, segmentation and per-pixel attribution, rarely examined parameters, have a significant impact. The implementation of and data gathered in this work are available online: https://github.com/guspih/post-hoc-image-perturbation and https://bit.ly/smooth-mask-perturbation.
URLs: https://github.com/guspih/post-hoc-image-perturbation, https://bit.ly/smooth-mask-perturbation.
Authors: Dongdong Lin, Yue Li, Benedetta Tondi, Kaiqing Lin, Bin Li, Mauro Barni
Abstract: The rapid proliferation of Deep Neural Networks (DNNs) is driving a surge in model watermarking technologies, as the trained models themselves constitute valuable intellectual property. Existing watermarking approaches primarily focus on modifying model parameters or altering sampling behaviors. However, with the emergence of increasingly large models, improving the efficiency of watermark embedding becomes essential to manage increasing computational demands. Prioritizing efficiency not only optimizes resource utilization, making the watermarking process more applicable for large models, but also mitigates potential degradation of model performance. In this paper, we propose an efficient watermarking method for Latent Diffusion Models (LDMs) based on Low-Rank Adaptation (LoRA). The core idea is to introduce trainable low-rank parameters into the frozen LDM to embed watermark, thereby preserving the integrity of the original model weights. Furthermore, a dynamic loss weight scheduler is designed to adaptively balance the objectives of generative quality and watermark fidelity, enabling the model to achieve effective watermark embedding with minimal impact on quality of the generated images. Experimental results show that the proposed method ensures fast and accurate watermark embedding and a high quality of the generated images, at the same time maintaining a level of robustness aligned - in some cases superior - with state-of-the-art approaches. Moreover, the method generalizes well across different datasets and base LDMs. Codes are available at: https://github.com/MrDongdongLin/EW-LoRA.
Authors: Camilo Espinosa-Curilem, Millaray Curilem, Daniel Basualto
Abstract: In volcano monitoring, effective recognition of seismic events is essential for understanding volcanic activity and raising timely warning alerts. Traditional methods rely on manual analysis, which can be subjective and labor-intensive. Furthermore, current automatic approaches often tackle detection and classification separately, mostly rely on single station information and generally require tailored preprocessing and representations to perform predictions. These limitations often hinder their application to real-time monitoring and utilization across different volcano conditions. This study introduces a novel approach that utilizes Semantic Segmentation models to automate seismic event recognition by applying a straight forward transformation of multi-channel 1D signals into 2D representations, enabling their use as images. Our framework employs a data-driven, end-to-end design that integrates multi-station seismic data with minimal preprocessing, performing both detection and classification simultaneously for five seismic event classes. We evaluated four state-of-the-art segmentation models (UNet, UNet++, DeepLabV3+ and SwinUNet) on approximately 25.000 seismic events recorded at four different Chilean volcanoes: Nevados del Chill\'an Volcanic Complex, Laguna del Maule, Villarrica and Puyehue-Cord\'on Caulle. Among these models, the UNet architecture was identified as the most effective model, achieving mean F1 and Intersection over Union (IoU) scores of up to 0.91 and 0.88, respectively, and demonstrating superior noise robustness and model flexibility to unseen volcano datasets.
Authors: Yifan Wang, Xiaochen Yang, Fanqi Pu, Qingmin Liao, Wenming Yang
Abstract: Monocular 3D object detection has attracted great attention due to simplicity and low cost. Existing methods typically follow conventional 2D detection paradigms, first locating object centers and then predicting 3D attributes via neighboring features. However, these methods predominantly rely on progressive cross-scale feature aggregation and focus solely on local information, which may result in a lack of global awareness and the omission of small-scale objects. In addition, due to large variation in object scales across different scenes and depths, inaccurate receptive fields often lead to background noise and degraded feature representation. To address these issues, we introduces MonoASRH, a novel monocular 3D detection framework composed of Efficient Hybrid Feature Aggregation Module (EH-FAM) and Adaptive Scale-Aware 3D Regression Head (ASRH). Specifically, EH-FAM employs multi-head attention with a global receptive field to extract semantic features for small-scale objects and leverages lightweight convolutional modules to efficiently aggregate visual features across different scales. The ASRH encodes 2D bounding box dimensions and then fuses scale features with the semantic features aggregated by EH-FAM through a scale-semantic feature fusion module. The scale-semantic feature fusion module guides ASRH in learning dynamic receptive field offsets, incorporating scale priors into 3D position prediction for better scale-awareness. Extensive experiments on the KITTI and Waymo datasets demonstrate that MonoASRH achieves state-of-the-art performance.
Authors: Yuhan Pan, Yanan Sun, Wei Gong
Abstract: Long-Tailed (LT) recognition has been widely studied to tackle the challenge of imbalanced data distributions in real-world applications. However, the design of neural architectures for LT settings has received limited attention, despite evidence showing that architecture choices can substantially affect performance. This paper aims to bridge the gap between LT challenges and neural network design by providing an in-depth analysis of how various architectures influence LT performance. Specifically, we systematically examine the effects of key network components on LT handling, such as topology, convolutions, and activation functions. Based on these observations, we propose two convolutional operations optimized for improved performance. Recognizing that operation interactions are also crucial to network effectiveness, we apply Neural Architecture Search (NAS) to facilitate efficient exploration. We propose LT-DARTS, a NAS method with a novel search space and search strategy specifically designed for LT data. Experimental results demonstrate that our approach consistently outperforms existing architectures across multiple LT datasets, achieving parameter-efficient, state-of-the-art results when integrated with current LT methods.
Authors: Evelyn J. Mannix, Liam Hodgkinson, Howard Bondell
Abstract: Knowledge distillation methods compress models by training a student network using the classification outputs of a high quality teacher model, but can fail to effectively transfer the properties of computer vision foundation models from the teacher to the student. While it has been recently shown that feature distillation$\unicode{x2013}$where a teacher model's output features are replicated instead$\unicode{x2013}$can reproduce performance for foundation models across numerous downstream tasks, they fall short in matching critical properties such as robustness and out-of-distribution (OOD) detection performance. This paper overcomes this shortcoming by introducing Cosine-similarity Preserving Compression (CosPress), a feature distillation technique that learns a mapping to compress the latent space of the teacher model into the smaller latent space of the student, by preserving the cosine similarities between image embeddings. This enables direct optimisation of the student network and produces a more faithful reproduction of the teacher's properties. It is shown that distillation with CosPress on a variety of datasets, including ImageNet, produces more accurate models with greater performance on generalisability, robustness and OOD detection benchmarks, and that this technique provides a competitive pathway for training highly performant lightweight models on small datasets. Code is available at github.com/emannix/cospress.
Authors: Yuhang Han, Xuyang Liu, Zihan Zhang, Pengxiang Ding, Junjie Chen, Donglin Wang, Honggang Chen, Qingsen Yan, Siteng Huang
Abstract: The quadratic complexity of Multimodal Large Language Models (MLLMs) with respect to context length poses significant computational and memory challenges, hindering their real-world deployment. In the paper, we devise a ''filter-correlate-compress'' framework to accelerate the MLLM by systematically optimizing multimodal context length during prefilling. The framework first implements FiCoCo-V, a training-free method operating within the vision encoder. It employs a redundancy-based token discard mechanism that uses a novel integrated metric to accurately filter out redundant visual tokens. To mitigate information loss, the framework introduces a correlation-based information recycling mechanism that allows preserved tokens to selectively recycle information from correlated discarded tokens with a self-preserving compression, thereby preventing the dilution of their own core content. The framework's FiCoCo-L variant further leverages task-aware textual priors to perform token reduction directly within the LLM decoder. Extensive experiments demonstrate that the FiCoCo series effectively accelerates a range of MLLMs, achieves up to 14.7x FLOPs reduction with 93.6% performance retention. Our methods consistently outperform state-of-the-art training-free approaches, showcasing effectiveness and generalizability across model architectures, sizes, and tasks without requiring retraining. Code: https://github.com/kawhiiiileo/FiCoCo
Authors: Chade Li, Pengju Zhang, Jiaming Zhang, Yihong Wu
Abstract: 3D point cloud segmentation has a wide range of applications in areas such as autonomous driving, augmented reality, virtual reality and digital twins. The point cloud data collected in real scenes often contain small objects and categories with small sample sizes, which are difficult to handle by existing networks. In this regard, we propose a point cloud segmentation network that fuses local attention based on density perception with global attention. The core idea is to increase the effective receptive field of each point while reducing the loss of information about small objects in dense areas. Specifically, we divide different sized windows for local areas with different densities to compute attention within the window. Furthermore, we consider each local area as an independent token for the global attention of the entire input. A category-response loss is also proposed to balance the processing of different categories and sizes of objects. In particular, we set up an additional fully connected layer in the middle of the network for prediction of the presence of object categories, and construct a binary cross-entropy loss to respond to the presence of categories in the scene. In experiments, our method achieves competitive results in semantic segmentation and part segmentation tasks on several publicly available datasets. Experiments on point cloud data obtained from complex real-world scenes filled with tiny objects also validate the strong segmentation capability of our method for small objects as well as small sample categories.
Authors: Yujin Wang, Quanfeng Liu, Jiaqi Fan, Jinlong Hong, Hongqing Chu, Mengjian Tian, Bingzhao Gao, Hong Chen
Abstract: Understanding and addressing corner cases is essential for ensuring the safety and reliability of autonomous driving systems. Vision-language models (VLMs) play a crucial role in enhancing scenario comprehension, yet they face significant challenges, such as hallucination and insufficient real-world grounding, which compromise their performance in critical driving scenarios. In this work, RAC3, a novel framework designed to enhance the performance of VLMs in corner case comprehension, is proposed. RAC3 integrates a frequency-spatial fusion (FSF) image encoder, a cross-modal alignment training method for embedding models with hard and semi-hard negative mining, and a fast querying and retrieval pipeline based on K-Means clustering and hierarchical navigable small world (HNSW) indexing. A multimodal chain-of-thought (CoT) prompting strategy to guide analogical reasoning and reduce hallucinations during inference is introduced. Moreover, an update mechanism is integrated into RAC3 to ensure continual learning within the framework. Extensive experiments on the CODA and nuScenes datasets demonstrate that RAC3 significantly improves corner case comprehension across multiple downstream tasks. Compared to prior state-of-the-art methods, RAC3 achieves the highest final score of 74.46 on the CODA-LM benchmark and shows consistent performance gains when integrated with end-to-end frameworks like DriveLM. These results demonstrate the effectiveness of retrieval-augmented strategies and cross-modal alignment for safer and more interpretable autonomous driving.
Authors: Muhammet Esat Kalfaoglu, Halil Ibrahim Ozturk, Ozsel Kilinc, Alptekin Temizel
Abstract: Understanding road topology is crucial for autonomous driving. This paper introduces TopoBDA (Topology with Bezier Deformable Attention), a novel approach that enhances road topology comprehension by leveraging Bezier Deformable Attention (BDA). TopoBDA processes multi-camera 360-degree imagery to generate Bird's Eye View (BEV) features, which are refined through a transformer decoder employing BDA. BDA utilizes Bezier control points to drive the deformable attention mechanism, improving the detection and representation of elongated and thin polyline structures, such as lane centerlines. Additionally, TopoBDA integrates two auxiliary components: an instance mask formulation loss and a one-to-many set prediction loss strategy, to further refine centerline detection and enhance road topology understanding. Experimental evaluations on the OpenLane-V2 dataset demonstrate that TopoBDA outperforms existing methods, achieving state-of-the-art results in centerline detection and topology reasoning. TopoBDA also achieves the best results on the OpenLane-V1 dataset in 3D lane detection. Further experiments on integrating multi-modal data -- such as LiDAR, radar, and SDMap -- show that multimodal inputs can further enhance performance in road topology understanding.
Authors: Linhao Huang, Xue Jiang, Zhiqiang Wang, Wentao Mo, Xi Xiao, Bo Han, Yongjie Yin, Feng Zheng
Abstract: Video-based multimodal large language models (V-MLLMs) have shown vulnerability to adversarial examples in video-text multimodal tasks. However, the transferability of adversarial videos to unseen models - a common and practical real-world scenario - remains unexplored. In this paper, we pioneer an investigation into the transferability of adversarial video samples across V-MLLMs. We find that existing adversarial attack methods face significant limitations when applied in black-box settings for V-MLLMs, which we attribute to the following shortcomings: (1) lacking generalization in perturbing video features, (2) focusing only on sparse key-frames, and (3) failing to integrate multimodal information. To address these limitations and deepen the understanding of V-MLLM vulnerabilities in black-box scenarios, we introduce the Image-to-Video MLLM (I2V-MLLM) attack. In I2V-MLLM, we utilize an image-based multimodal large language model (I-MLLM) as a surrogate model to craft adversarial video samples. Multimodal interactions and spatiotemporal information are integrated to disrupt video representations within the latent space, improving adversarial transferability. Additionally, a perturbation propagation technique is introduced to handle different unknown frame sampling strategies. Experimental results demonstrate that our method can generate adversarial examples that exhibit strong transferability across different V-MLLMs on multiple video-text multimodal tasks. Compared to white-box attacks on these models, our black-box attacks (using BLIP-2 as a surrogate model) achieve competitive performance, with average attack success rate (AASR) of 57.98% on MSVD-QA and 58.26% on MSRVTT-QA for Zero-Shot VideoQA tasks, respectively.
Authors: Jinze Yu, Yiqun Wang, Aiheng Jiang, Zhengda Lu, Jianwei Guo, Yong Li, Hongxing Qin, Xiaopeng Zhang
Abstract: Current novel view synthesis methods are typically designed for high-quality and clean input images. However, in foggy scenes, scattering and attenuation can significantly degrade the quality of rendering. Although NeRF-based dehazing approaches have been developed, their reliance on deep fully connected neural networks and per-ray sampling strategies leads to high computational costs. Furthermore, NeRF's implicit representation limits its ability to recover fine-grained details from hazy scenes. To overcome these limitations, we propose learning an explicit Gaussian representation to explain the formation mechanism of foggy images through a physically forward rendering process. Our method, DehazeGS, reconstructs and renders fog-free scenes using only multi-view foggy images as input. Specifically, based on the atmospheric scattering model, we simulate the formation of fog by establishing the transmission function directly onto Gaussian primitives via depth-to-transmission mapping. During training, we jointly learn the atmospheric light and scattering coefficients while optimizing the Gaussian representation of foggy scenes. At inference time, we remove the effects of scattering and attenuation in Gaussian distributions and directly render the scene to obtain dehazed views. Experiments on both real-world and synthetic foggy datasets demonstrate that DehazeGS achieves state-of-the-art performance. visualizations are available at https://dehazegs.github.io/
Authors: Tinglei Wan, Tonghua Su, Zhongjie Wang
Abstract: Structured light (SL) 3D reconstruction captures the precise surface shape of objects, providing high-accuracy 3D data essential for industrial inspection and cultural heritage digitization. However, existing methods suffer from two key limitations: reliance on scene-specific calibration with manual parameter tuning, and optimization frameworks tailored to specific SL patterns, limiting their generalizability across varied scenarios. We propose General and Unified Structured Light Optimization (GUSLO), a novel framework addressing these issues through two coordinated innovations: (1) single-shot calibration via 2D triangulation-based interpolation that converts sparse matches into dense correspondence fields, and (2) artifact-aware photometric adaptation via explicit transfer functions, balancing generalization and color fidelity. We conduct diverse experiments covering binary, speckle, and color-coded settings. Results show that GUSLO consistently improves accuracy and cross-encoding robustness over conventional methods in challenging industrial and cultural scenarios.
Authors: Zhenyu Yang, Yuhang Hu, Zemin Du, Dizhan Xue, Shengsheng Qian, Jiahong Wu, Fan Yang, Weiming Dong, Changsheng Xu
Abstract: Despite the significant advancements of Large Vision-Language Models (LVLMs) on established benchmarks, there remains a notable gap in suitable evaluation regarding their applicability in the emerging domain of long-context streaming video understanding. Current benchmarks for video understanding typically emphasize isolated single-instance text inputs and fail to evaluate the capacity to sustain temporal reasoning throughout the entire duration of video streams. To address these limitations, we introduce SVBench, a pioneering benchmark with temporal multi-turn question-answering chains specifically designed to thoroughly assess the capabilities of streaming video understanding of current LVLMs. We design a semi-automated annotation pipeline to obtain 49,979 Question-Answer (QA) pairs of 1,353 streaming videos, which includes generating QA chains that represent a series of consecutive multi-turn dialogues over video segments and constructing temporal linkages between successive QA chains. Our experimental results, obtained from 14 models in dialogue and streaming evaluations, reveal that while the closed-source GPT-4o outperforms others, most open-source LVLMs struggle with long-context streaming video understanding. We also construct a StreamingChat model, which significantly outperforms open-source LVLMs on our SVBench and achieves comparable performance on diverse vision-language benchmarks. We expect SVBench to advance the research of streaming video understanding by providing a comprehensive and in-depth analysis of current LVLMs. Our benchmark and model can be accessed at https://github.com/sotayang/SVBench.
Authors: Dariush Lotfi, Mohammad-Ali Nikouei Mahani, Mohamad Koohi-Moghadam, Kyongtae Ty Bae
Abstract: In AI-driven medical imaging, the failure to detect out-of-distribution (OOD) data poses a severe risk to clinical reliability, potentially leading to critical diagnostic errors. Current OOD detection methods often demand impractical retraining or modifications to pre-trained models, hindering their adoption in regulated clinical environments. To address this challenge, we propose a post-hoc normalizing flow-based approach that seamlessly integrates with existing pre-trained models without altering their weights. We evaluate the approach on our in-house-curated MedOOD dataset, designed to capture clinically relevant distribution shifts, and on the MedMNIST benchmark. The proposed method achieves an AUROC of 84.61% on MedOOD, outperforming ViM (80.65%) and MDS (80.87%), and reaches 93.8% AUROC on MedMNIST, surpassing ViM (88.08%) and ReAct (87.05%). This combination of strong performance and post-hoc integration capability makes our approach a practical and effective safeguard for clinical imaging workflows. The model and code to build OOD datasets are publicly accessible at https://github.com/dlotfi/MedOODFlow.
Authors: Chi Ruan, Jiying Zhao, Wenhu Chen
Abstract: Although open-vocabulary object detectors can generalize to unseen categories, they still rely on predefined textual prompts or classifier heads during inference. Recent generative object detectors address this limitation by coupling an autoregressive language model with a detector backbone, enabling direct category name generation for each detected object. However, this straightforward design introduces structural redundancy and substantial latency. In this paper, we propose a Real-Time Generative Detection Transformer (RTGen), a real-time generative object detector with a succinct encoder-decoder architecture. Specifically, we introduce a novel Region-Language Decoder (RL-Decoder) that jointly decodes visual and textual representations within a unified framework. The textual side is organized as a Directed Acyclic Graph (DAG), enabling non-autoregressive category naming. Benefiting from these designs, RTGen-R34 achieves 131.3 FPS on T4 GPUs, over 270x faster than GenerateU. Moreover, our models learn to generate category names directly from detection labels, without relying on external supervision such as CLIP or pretrained language models, achieving efficient and flexible open-ended detection.
Authors: Tianyi Wang, Jianan Fan, Dingxin Zhang, Dongnan Liu, Yong Xia, Heng Huang, Weidong Cai
Abstract: Histopathology and transcriptomics are fundamental modalities in oncology, encapsulating the morphological and molecular aspects of the disease. Multi-modal self-supervised learning has demonstrated remarkable potential in learning pathological representations by integrating diverse data sources. Conventional multi-modal integration methods primarily emphasize modality alignment, while paying insufficient attention to retaining the modality-specific structures. However, unlike conventional scenarios where multi-modal inputs share highly overlapping features, histopathology and transcriptomics exhibit pronounced heterogeneity, offering orthogonal yet complementary insights. Histopathology provides morphological and spatial context, elucidating tissue architecture and cellular topology, whereas transcriptomics delineates molecular signatures through gene expression patterns. This inherent disparity introduces a major challenge in aligning them while maintaining modality-specific fidelity. To address these challenges, we present MIRROR, a novel multi-modal representation learning method designed to foster both modality alignment and retention. MIRROR employs dedicated encoders to extract comprehensive features for each modality, which is further complemented by a modality alignment module to achieve seamless integration between phenotype patterns and molecular profiles. Furthermore, a modality retention module safeguards unique attributes from each modality, while a style clustering module mitigates redundancy and enhances disease-relevant information by modeling and aligning consistent pathological signatures within a clustering space. Extensive evaluations on TCGA cohorts for cancer subtyping and survival analysis highlight MIRROR's superior performance, demonstrating its effectiveness in constructing comprehensive oncological feature representations and benefiting the cancer diagnosis.
Authors: Yuyan Chen, Nico Lang, B. Christian Schmidt, Aditya Jain, Yves Basset, Sara Beery, Maxim Larriv\'ee, David Rolnick
Abstract: Global biodiversity is declining at an unprecedented rate, yet little information is known about most species and how their populations are changing. Indeed, some 90% of Earth's species are estimated to be completely unknown. Machine learning has recently emerged as a promising tool to facilitate long-term, large-scale biodiversity monitoring, including algorithms for fine-grained classification of species from images. However, such algorithms typically are not designed to detect examples from categories unseen during training -- the problem of open-set recognition (OSR) -- limiting their applicability for highly diverse, poorly studied taxa such as insects. To address this gap, we introduce Open-Insect, a large-scale, fine-grained dataset to evaluate unknown species detection across different geographic regions with varying difficulty. We benchmark 38 OSR algorithms across three categories: post-hoc, training-time regularization, and training with auxiliary data, finding that simple post-hoc approaches remain a strong baseline. We also demonstrate how to leverage auxiliary data to improve species discovery in regions with limited data. Our results provide insights to guide the development of computer vision methods for biodiversity monitoring and species discovery.
Authors: Gangwei Xu, Haotong Lin, Zhaoxing Zhang, Hongcheng Luo, Haiyang Sun, Xin Yang
Abstract: Event cameras deliver visual information characterized by a high dynamic range and high temporal resolution, offering significant advantages in estimating optical flow for complex lighting conditions and fast-moving objects. Current advanced optical flow methods for event cameras largely adopt established image-based frameworks. However, the spatial sparsity of event data limits their performance. In this paper, we present BAT, an innovative framework that estimates event-based optical flow using bidirectional adaptive temporal correlation. BAT includes three novel designs: 1) a bidirectional temporal correlation that transforms bidirectional temporally dense motion cues into spatially dense ones, enabling accurate and spatially dense optical flow estimation; 2) an adaptive temporal sampling strategy for maintaining temporal consistency in correlation; 3) spatially adaptive temporal motion aggregation to efficiently and adaptively aggregate consistent target motion features into adjacent motion features while suppressing inconsistent ones. Our results rank $1^{st}$ on the DSEC-Flow benchmark, outperforming existing state-of-the-art methods by a large margin while also exhibiting sharp edges and high-quality details. Notably, our BAT can accurately predict future optical flow using only past events, significantly outperforming E-RAFT's warm-start approach. Code: \textcolor{magenta}{https://github.com/gangweiX/BAT}.
Authors: Adrien Meyer, Lorenzo Arboit, Giuseppe Massimiani, Shih-Min Yin, Didier Mutter, Nicolas Padoy
Abstract: Purpose: The Segment Anything Model (SAM) promises to ease the annotation bottleneck in medical segmentation, but overlapping anatomy and blurred boundaries make its point prompts ambiguous, leading to cycles of manual refinement to achieve precise masks. Better prompting strategies are needed. Methods: We propose a structured prompting strategy using 4 points as a compact instance-level shape description. We study two 4-point variants: extreme points and the proposed major/minor axis endpoints, inspired by ultrasound measurement practice. SAM cannot fully exploit such structured prompts because it treats all points identically and lacks geometry-aware reasoning. To address this, we introduce S4M (4-points to Segment Anything), which augments SAM to interpret 4 points as relational cues rather than isolated clicks. S4M expands the prompt space with role-specific embeddings and adds an auxiliary "Canvas" pretext task that sketches coarse masks directly from prompts, fostering geometry-aware reasoning. Results: Across eight datasets in ultrasound and surgical endoscopy, S4M improves segmentation by +3.42 mIoU over a strong SAM baseline at equal prompt budget. An annotation study with three clinicians further shows that major/minor prompts enable faster annotation. Conclusion: S4M increases performance, reduces annotation effort, and aligns prompting with clinical practice, enabling more scalable dataset development in medical imaging.
Authors: Jing Zhang, Zhikai Li, Chengzhi Hu, Xuewen Liu, Qingyi Gu
Abstract: Segment Anything Model (SAM) exhibits remarkable zero-shot segmentation capability; however, its prohibitive computational costs make edge deployment challenging. Although post-training quantization (PTQ) offers a promising compression solution, existing methods yield unsatisfactory results when applied to SAM, owing to its specialized model components and promptable workflow: (i) The mask decoder's attention exhibits extreme activation outliers, and we find that aggressive clipping (even 100x), without smoothing or isolation, is effective in suppressing outliers while maintaining performance. Unfortunately, traditional distribution-based metrics (e.g., MSE) fail to provide such large-scale clipping. (ii) Existing quantization reconstruction methods neglect semantic interactivity of SAM, leading to misalignment between image feature and prompt intention. To address the above issues, we propose SAQ-SAM in this paper, which boosts PTQ for SAM from the perspective of semantic alignment. Specifically, we propose Perceptual-Consistency Clipping, which exploits attention focus overlap to promote aggressive clipping while preserving semantic capabilities. Furthermore, we propose Prompt-Aware Reconstruction, which incorporates image-prompt interactions by leveraging cross-attention in mask decoder, thus facilitating alignment in both distribution and semantic. Moreover, to ensure the interaction efficiency, we design a layer-skipping strategy for image tokens in encoder. Extensive experiments are conducted on various SAM sizes and tasks, including instance segmentation, oriented object detection, and semantic segmentation, and the results show that our method consistently exhibits advantages. For example, when quantizing SAM-B to 4-bit, SAQ-SAM achieves 11.7% higher mAP than the baseline in instance segmentation task.
Authors: Haiyang Xie, Xi Shen, Shihua Huang, Qirui Wang, Zheng Wang
Abstract: Most visual models are designed for sRGB images, yet RAW data offers significant advantages for object detection by preserving sensor information before ISP processing. This enables improved detection accuracy and more efficient hardware designs by bypassing the ISP. However, RAW object detection is challenging due to limited training data, unbalanced pixel distributions, and sensor noise. To address this, we propose SimROD, a lightweight and effective approach for RAW object detection. We introduce a Global Gamma Enhancement (GGE) module, which applies a learnable global gamma transformation with only four parameters, improving feature representation while keeping the model efficient. Additionally, we leverage the green channel's richer signal to enhance local details, aligning with the human eye's sensitivity and Bayer filter design. Extensive experiments on multiple RAW object detection datasets and detectors demonstrate that SimROD outperforms state-of-the-art methods like RAW-Adapter and DIAP while maintaining efficiency. Our work highlights the potential of RAW data for real-world object detection. Code is available at https://ocean146.github.io/SimROD2025/.
Authors: Yehyun Suh, J. Ryan Martin, Daniel Moyer
Abstract: This paper presents an approach for improving 2D/3D pelvis registration in optimization-based pose estimators using a learned initialization function. Current methods often fail to converge to the optimal solution when initialized naively. We find that even a coarse initializer greatly improves pose estimator accuracy, and improves overall computational efficiency. This approach proves to be effective also in challenging cases under more extreme pose variation. Experimental validation demonstrates that our method consistently achieves robust and accurate registration, enhancing the reliability of 2D/3D registration for clinical applications.
Authors: Xinghan Li, Yue Yu, Xue Song, Haijun Shan, Jingjing Chen
Abstract: With the rapid advancement of vision generation models, the potential security risks stemming from synthetic visual content have garnered increasing attention, posing significant challenges for AI-generated image detection. Existing methods suffer from inadequate generalization capabilities, resulting in unsatisfactory performance on emerging generative models. To address this issue, this paper presents NIRNet (Noise-based Imprint Revealing Network), a novel framework that leverages noise-based imprint for the detection task. Specifically, we propose a novel Noise-based Imprint Simulator to capture intrinsic patterns imprinted in images generated by different models. By aggregating imprint from various generative models, imprint of future models can be extrapolated to expand training data, thereby enhancing generalization and robustness. Furthermore, we design a new pipeline that pioneers the use of noise patterns, derived from a Noise-based Imprint Extractor, alongside other visual features for AI-generated image detection, significantly improving detection performance. Our approach achieves state-of-the-art performance across seven diverse benchmarks, including five public datasets and two newly proposed generalization tests, demonstrating its superior generalization and effectiveness. Paper Submission: pdf
Authors: Donglin Yang, Paul Vicol, Xiaojuan Qi, Renjie Liao, Xiaofan Zhang
Abstract: Deep learning-based super-resolution (SR) methods often perform pixel-wise computations uniformly across entire images, even in homogeneous regions where high-resolution refinement is redundant. We propose the Quadtree Diffusion Model (QDM), a region-adaptive diffusion framework that leverages a quadtree structure to selectively enhance detail-rich regions while reducing computations in homogeneous areas. By guiding the diffusion with a quadtree derived from the low-quality input, QDM identifies key regions-represented by leaf nodes-where fine detail is essential and applies minimal refinement elsewhere. This mask-guided, two-stream architecture adaptively balances quality and efficiency, producing high-fidelity outputs with low computational redundancy. Experiments demonstrate QDM's effectiveness in high-resolution SR tasks across diverse image types, particularly in medical imaging (e.g., CT scans), where large homogeneous regions are prevalent. Furthermore, QDM outperforms or is comparable to state-of-the-art SR methods on standard benchmarks while significantly reducing computational costs, highlighting its efficiency and suitability for resource-limited environments. Our code is available at https://github.com/linYDTHU/QDM.
Authors: Oren Shrout, Ayellet Tal
Abstract: We propose SFMNet, a novel 3D sparse detector that combines the efficiency of sparse convolutions with the ability to model long-range dependencies. While traditional sparse convolution techniques efficiently capture local structures, they struggle with modeling long-range relationships. However, capturing long-range dependencies is fundamental for 3D object detection. In contrast, transformers are designed to capture these long-range dependencies through attention mechanisms. But, they come with high computational costs, due to their quadratic query-key-value interactions. Furthermore, directly applying attention to non-empty voxels is inefficient due to the sparse nature of 3D scenes. Our SFMNet is built on a novel Sparse Focal Modulation (SFM) module, which integrates short- and long-range contexts with linear complexity by leveraging a new hierarchical sparse convolution design. This approach enables SFMNet to achieve high detection performance with improved efficiency, making it well-suited for large-scale LiDAR scenes. We show that our detector achieves state-of-the-art performance on autonomous driving datasets.
Authors: Corentin Sautier, Gilles Puy, Alexandre Boulch, Renaud Marlet, Vincent Lepetit
Abstract: Panoptic segmentation of LiDAR point clouds is fundamental to outdoor scene understanding, with autonomous driving being a primary application. While state-of-the-art approaches typically rely on end-to-end deep learning architectures and extensive manual annotations of instances, the significant cost and time investment required for labeling large-scale point cloud datasets remains a major bottleneck in this field. In this work, we demonstrate that competitive panoptic segmentation can be achieved using only semantic labels, with instances predicted without any training or annotations. Our method outperforms {most} state-of-the-art supervised methods on standard benchmarks including SemanticKITTI and nuScenes, and outperforms every publicly available method on SemanticKITTI as a drop-in instance head replacement, while running in real-time on a single-threaded CPU and requiring no instance labels. It is fully explainable, and requires no learning or parameter tuning. Alpine combined with state-of-the-art semantic segmentation ranks first on the official panoptic segmentation leaderboard of SemanticKITTI. Code is available at https://github.com/valeoai/Alpine/
Authors: Shicheng Li, Lei Li, Kun Ouyang, Shuhuai Ren, Yuanxin Liu, Yuanxing Zhang, Fuzheng Zhang, Lingpeng Kong, Qi Liu, Xu Sun
Abstract: Video Large Language Models (Video LLMs) have achieved significant success by adopting the paradigm of large-scale pre-training followed by supervised fine-tuning (SFT). However, existing approaches struggle with temporal reasoning due to weak temporal correspondence in the data and over-reliance on the next-token prediction paradigm}, which collectively result in the absence temporal supervision. To address these limitations, we propose TEMPLE (TEMporal Preference LEarning), a systematic framework that enhances temporal reasoning capabilities through Direct Preference Optimization (DPO). To address temporal information scarcity in data, we introduce an automated pipeline for systematically constructing temporality-intensive preference pairs comprising three steps: selecting temporally rich videos, designing video-specific perturbation strategies, and evaluating model responses on clean and perturbed inputs. Complementing this data pipeline, we provide additional supervision signals via preference learning and propose a novel Progressive Pre-SFT Alignment strategy featuring two key innovations: a curriculum learning strategy which progressively increases perturbation difficulty to maximize data efficiency; and applying preference optimization before instruction tuning to incentivize fundamental temporal alignment. Extensive experiments demonstrate that our approach consistently improves Video LLM performance across multiple benchmarks with a relatively small set of self-generated DPO data. Our findings highlight TEMPLE as a scalable and efficient complement to SFT-based methods, paving the way for developing reliable Video LLMs.
Authors: Jichen Hu, Chen Yang, Zanwei Zhou, Jiemin Fang, Xiaokang Yang, Qi Tian, Wei Shen
Abstract: Reflection removal of a single image remains a highly challenging task due to the complex entanglement between target scenes and unwanted reflections. Despite significant progress, existing methods are hindered by the scarcity of high-quality, diverse data and insufficient restoration priors, resulting in limited generalization across various real-world scenarios. In this paper, we propose Dereflection Any Image, a comprehensive solution with an efficient data preparation pipeline and a generalizable model for robust reflection removal. First, we introduce a dataset named Diverse Reflection Removal (DRR) created by randomly rotating reflective mediums in target scenes, enabling variation of reflection angles and intensities, and setting a new benchmark in scale, quality, and diversity. Second, we propose a diffusion-based framework with one-step diffusion for deterministic outputs and fast inference. To ensure stable learning, we design a three-stage progressive training strategy, including reflection-invariant finetuning to encourage consistent outputs across varying reflection patterns that characterize our dataset. Extensive experiments show that our method achieves SOTA performance on both common benchmarks and challenging in-the-wild images, showing superior generalization across diverse real-world scenes.
Authors: Zexu Huang, Min Xu, Stuart Perry
Abstract: Recent developments in 3D reconstruction and neural rendering have significantly propelled the capabilities of photo-realistic 3D scene rendering across various academic and industrial fields. The 3D Gaussian Splatting technique, alongside its derivatives, integrates the advantages of primitive-based and volumetric representations to deliver top-tier rendering quality and efficiency. Despite these advancements, the method tends to generate excessive redundant noisy Gaussians overfitted to every training view, which degrades the rendering quality. Additionally, while 3D Gaussian Splatting excels in small-scale and object-centric scenes, its application to larger scenes is hindered by constraints such as limited video memory, excessive optimization duration, and variable appearance across views. To address these challenges, we introduce GaussianFocus, an innovative approach that incorporates a patch attention algorithm to refine rendering quality and implements a Gaussian constraints strategy to minimize redundancy. Moreover, we propose a subdivision reconstruction strategy for large-scale scenes, dividing them into smaller, manageable blocks for individual training. Our results indicate that GaussianFocus significantly reduces unnecessary Gaussians and enhances rendering quality, surpassing existing State-of-The-Art (SoTA) methods. Furthermore, we demonstrate the capability of our approach to effectively manage and render large scenes, such as urban environments, whilst maintaining high fidelity in the visual output.
Authors: Olgica Zaric, Carmen Leser, Vladimir Juras, Alex Farr, Pavol Szomolanyi, Malina Gologan, Stanislas Rapacchi, Laura Villazan Garcia, Haider Ali, Christian Singer, Siegfried Trattnig, Christian Licht, Ramona Woitek
Abstract: Introduction: In sodium (23Na) magnetic resonance imaging (MRI), partial volume effects (PVE) are one of the most common causes of errors in the in vivo quantification of tissue sodium concentration (TSC). Advanced image reconstruction algorithms, such as compressed sensing (CS), have the potential to reduce PVE. Therefore, we investigated the feasibility of using CS-based methods to improve image quality and TSC quantification accuracy in patients with breast cancer. Subjects and methods: In this study, three healthy participants and 12 female participants with breast cancer were examined on a 7T MRI scanner. 23Na-MRI images were reconstructed using weighted total variation (wTV), directional total variation (dTV), anatomically guided total variation (AG-TV) and adaptive combine (ADC) methods. The consistency of tumor volume delineations based on sodium data was assessed using the Dice score, and TSC quantification was performed for various image reconstruction methods. Pearsons correlation coefficients were calculated to assess the relationships between wTV, dTV, AG-TV, and ADC values. Results: All methods provided breast MRI images with well-preserved sodium signal and tissue structures. The mean Dice scores for wTV, dTV, and AG-TV were 65%, 72%, and 75%, respectively. Average TSC values in breast tumors were 61.0, 72.0, 73.0, and 88.0 mmol/L for wTV, dTV, AG-TV, and ADC, respectively. A strong negative correlation was observed between wTV and dTV (r = -0.78, 95% CI [-0.94, -0.31], p = 0.0076) and a strong positive correlation between dTV and AG-TV (r = 0.71, 95% CI [0.16, 0.92], p = 0.0207) was found. Conclusion: The results of this study showed that differences in tumor appearance and TSC estimations may depend on the type of image reconstruction and the parameters used. This is most likely due to differences in their ability to reduce PVE.
Authors: Yuli Zhou, Yawei Li, Yuqian Fu, Luca Benini, Ender Konukoglu, Guolei Sun
Abstract: Video camouflaged object segmentation (VCOS), aiming at segmenting camouflaged objects that seamlessly blend into their environment, is a fundamental vision task with various real-world applications. With the release of SAM2, video segmentation has witnessed significant progress. However, SAM2's capability of segmenting camouflaged videos is suboptimal, especially when given simple prompts such as point and box. To address the problem, we propose Camouflaged SAM2 (CamSAM2), which enhances SAM2's ability to handle camouflaged scenes without modifying SAM2's parameters. Specifically, we introduce a decamouflaged token to provide the flexibility of feature adjustment for VCOS. To make full use of fine-grained and high-resolution features from the current frame and previous frames, we propose implicit object-aware fusion (IOF) and explicit object-aware fusion (EOF) modules, respectively. Object prototype generation (OPG) is introduced to abstract and memorize object prototypes with informative details using high-quality features from previous frames. Extensive experiments are conducted to validate the effectiveness of our approach. While CamSAM2 only adds negligible learnable parameters to SAM2, it substantially outperforms SAM2 on three VCOS datasets, especially achieving 12.2 mDice gains with click prompt on MoCA-Mask and 19.6 mDice gains with mask prompt on SUN-SEG-Hard, with Hiera-T as the backbone. The code is available at https://github.com/zhoustan/CamSAM2.
Authors: Carlos Plou, Cesar Borja, Ruben Martinez-Cantin, Ana C. Murillo
Abstract: Finding information in hour-long videos is a challenging task even for top-performing Vision Language Models (VLMs), as encoding visual content quickly exceeds available context windows. To tackle this challenge, we present FALCONEye, a novel video agent based on a training-free, model-agnostic meta-architecture composed of a VLM and a Large Language Model (LLM). FALCONEye answers open-ended questions using an exploration-based search algorithm guided by calibrated confidence from the VLM's answers. We also introduce the FALCON-Bench benchmark, extending Question Answering problem to Video Answer Search-requiring models to return both the answer and its supporting temporal window for open-ended questions in hour-long videos. With just a 7B VLM and a lightweight LLM, FALCONEye outscores all open-source 7B VLMs and comparable agents in FALCON-Bench. It further demonstrates its generalization capability in MLVU benchmark with shorter videos and different tasks, surpassing GPT-4o on single-detail tasks while slashing inference cost by roughly an order of magnitude.
Authors: Yunusa Haruna, Adamu Lawan
Abstract: Capturing long-range dependencies efficiently is essential for visual recognition tasks, yet existing methods face limitations. Convolutional neural networks (CNNs) struggle with restricted receptive fields, while Vision Transformers (ViTs) achieve global context and long-range modeling at a high computational cost. State-space models (SSMs) offer an alternative, but their application in vision remains underexplored. This work introduces vGamba, a hybrid vision backbone that integrates SSMs with attention mechanisms to enhance efficiency and expressiveness. At its core, the Gamba bottleneck block that includes, Gamba Cell, an adaptation of Mamba for 2D spatial structures, alongside a Multi-Head Self-Attention (MHSA) mechanism and a Gated Fusion Module for effective feature representation. The interplay of these components ensures that vGamba leverages the low computational demands of SSMs while maintaining the accuracy of attention mechanisms for modeling long-range dependencies in vision tasks. Additionally, the Fusion module enables seamless interaction between these components. Extensive experiments on classification, detection, and segmentation tasks demonstrate that vGamba achieves a superior trade-off between accuracy and computational efficiency, outperforming several existing models.
Authors: Haiduo Huang, Yadong Zhang, Yinghui Xu, Pengju Ren
Abstract: Dynamic convolution enhances model capacity by adaptively combining multiple kernels, yet faces critical trade-offs: prior works either (1) incur significant parameter overhead by scaling kernel numbers linearly, (2) compromise inference speed through complex kernel interactions, or (3) struggle to jointly optimize dynamic attention and static kernels. We observe that pre-trained Convolutional Neural Networks (CNNs) exhibit inter-layer redundancy akin to that in Large Language Models (LLMs). Specifically, dense convolutional layers can be efficiently replaced by derived "child" layers generated from a shared "parent" convolutional kernel through an adapter. To address these limitations and implement the weight-sharing mechanism, we propose a lightweight convolution kernel plug-in, named KernelDNA. It decouples kernel adaptation into input-dependent dynamic routing and pre-trained static modulation, ensuring both parameter efficiency and hardware-friendly inference. Unlike existing dynamic convolutions that expand parameters via multi-kernel ensembles, our method leverages cross-layer weight sharing and adapter-based modulation, enabling dynamic kernel specialization without altering the standard convolution structure. This design preserves the native computational efficiency of standard convolutions while enhancing representation power through input-adaptive kernel adjustments. Experiments on image classification and dense prediction tasks demonstrate that KernelDNA achieves a state-of-the-art accuracy-efficiency balance among dynamic convolution variants.
Authors: Jongseo Lee, Joohyun Chang, Dongho Lee, Jinwoo Choi
Abstract: We propose Cross-Attention in Audio, Space, and Time (CA^2ST), a transformer-based method for holistic video recognition. Recognizing actions in videos requires both spatial and temporal understanding, yet most existing models lack a balanced spatio-temporal understanding of videos. To address this, we propose a novel two-stream architecture, called Cross-Attention in Space and Time (CAST), using only RGB input. In each layer of CAST, Bottleneck Cross-Attention (B-CA) enables spatial and temporal experts to exchange information and make synergistic predictions. For holistic video understanding, we extend CAST by integrating an audio expert, forming Cross-Attention in Visual and Audio (CAVA). We validate the CAST on benchmarks with different characteristics, EPIC-KITCHENS-100, Something-Something-V2, and Kinetics-400, consistently showing balanced performance. We also validate the CAVA on audio-visual action recognition benchmarks, including UCF-101, VGG-Sound, KineticsSound, and EPIC-SOUNDS. With a favorable performance of CAVA across these datasets, we demonstrate the effective information exchange among multiple experts within the B-CA module. In summary, CA^2ST combines CAST and CAVA by employing spatial, temporal, and audio experts through cross-attention, achieving balanced and holistic video understanding.
Authors: Abdelrahman Elskhawy, Mengze Li, Nassir Navab, Benjamin Busam
Abstract: In Scene Graph Generation (SGG), structured representations are extracted from visual inputs as object nodes and connecting predicates, enabling image-based reasoning for diverse downstream tasks. While fully supervised SGG has improved steadily, it suffers from training bias due to limited curated data and long-tail predicate distributions, leading to poor predicate diversity and degraded downstream performance. We present PRISM-0, a zero-shot open-vocabulary SGG framework that leverages foundation models in a bottom-up pipeline to capture a broad spectrum of predicates. Detected object pairs are filtered, described via a Vision-Language Model (VLM), and processed by a Large Language Model (LLM) to generate fine- and coarse-grained predicates, which are then validated by a Visual Question Answering (VQA) model. PRISM-0 modular, dataset-independent design enriches existing SGG datasets such as Visual Genome and produces diverse, unbiased graphs. While operating entirely in a zero-shot setting, PRISM-0 achieves performance on par with state-of-the-art weakly-supervised models on SGG benchmarks and even state-of-the-art supervised methods in tasks such as Sentence-to-Graph Retrieval.
Authors: Junjie Jiang, Zelin Wang, Manqi Zhao, Yin Li, DongSheng Jiang
Abstract: Inspired by Segment Anything 2, which generalizes segmentation from images to videos, we propose SAM2MOT--a novel segmentation-driven paradigm for multi-object tracking that breaks away from the conventional detection-association framework. In contrast to previous approaches that treat segmentation as auxiliary information, SAM2MOT places it at the heart of the tracking process, systematically tackling challenges like false positives and occlusions. Its effectiveness has been thoroughly validated on major MOT benchmarks. Furthermore, SAM2MOT integrates pre-trained detector, pre-trained segmentor with tracking logic into a zero-shot MOT system that requires no fine-tuning. This significantly reduces dependence on labeled data and paves the way for transitioning MOT research from task-specific solutions to general-purpose systems. Experiments on DanceTrack, UAVDT, and BDD100K show state-of-the-art results. Notably, SAM2MOT outperforms existing methods on DanceTrack by +2.1 HOTA and +4.5 IDF1, highlighting its effectiveness in MOT. Code is available at https://github.com/TripleJoy/SAM2MOT.
Authors: Jiani Liu, Zhiyuan Wang, Zeliang Zhang, Chao Huang, Susan Liang, Yunlong Tang, Chenliang Xu
Abstract: Vision Transformers (ViTs) have demonstrated impressive performance across a range of applications, including many safety-critical tasks. However, their unique architectural properties raise new challenges and opportunities in adversarial robustness. In particular, we observe that adversarial examples crafted on ViTs exhibit higher transferability compared to those crafted on CNNs, suggesting that ViTs contain structural characteristics favorable for transferable attacks. In this work, we investigate the role of computational redundancy in ViTs and its impact on adversarial transferability. Unlike prior studies that aim to reduce computation for efficiency, we propose to exploit this redundancy to improve the quality and transferability of adversarial examples. Through a detailed analysis, we identify two forms of redundancy, including the data-level and model-level, that can be harnessed to amplify attack effectiveness. Building on this insight, we design a suite of techniques, including attention sparsity manipulation, attention head permutation, clean token regularization, ghost MoE diversification, and test-time adversarial training. Extensive experiments on the ImageNet-1k dataset validate the effectiveness of our approach, showing that our methods significantly outperform existing baselines in both transferability and generality across diverse model architectures.
Authors: Dianbing Xi, Jiepeng Wang, Yuanzhi Liang, Xi Qiu, Yuchi Huo, Rui Wang, Chi Zhang, Xuelong Li
Abstract: In this paper, we propose a novel framework for controllable video diffusion, OmniVDiff , aiming to synthesize and comprehend multiple video visual content in a single diffusion model. To achieve this, OmniVDiff treats all video visual modalities in the color space to learn a joint distribution, while employing an adaptive control strategy that dynamically adjusts the role of each visual modality during the diffusion process, either as a generation modality or a conditioning modality. Our framework supports three key capabilities: (1) Text-conditioned video generation, where all modalities are jointly synthesized from a textual prompt; (2) Video understanding, where structural modalities are predicted from rgb inputs in a coherent manner; and (3) X-conditioned video generation, where video synthesis is guided by finegrained inputs such as depth, canny and segmentation. Extensive experiments demonstrate that OmniVDiff achieves state-of-the-art performance in video generation tasks and competitive results in video understanding. Its flexibility and scalability make it well-suited for downstream applications such as video-to-video translation, modality adaptation for visual tasks, and scene reconstruction.
Authors: Mahdi Alehdaghi, Rajarshi Bhattacharya, Pourya Shamsolmoali, Rafael M. O. Cruz, Eric Granger
Abstract: As AI systems grow more capable, it becomes increasingly important that their decisions remain understandable and aligned with human expectations. A key challenge is the limited interpretability of deep models. Post-hoc methods like GradCAM offer heatmaps but provide limited conceptual insight, while prototype-based approaches offer example-based explanations but often rely on rigid region selection and lack semantic consistency. To address these limitations, we propose PCMNet, a part-prototypical concept mining network that learns human-comprehensible prototypes from meaningful image regions without additional supervision. By clustering these prototypes into concept groups and extracting concept activation vectors, PCMNet provides structured, concept-level explanations and enhances robustness to occlusion and challenging conditions, which are both critical for building reliable and aligned AI systems. Experiments across multiple image classification benchmarks show that PCMNet outperforms state-of-the-art methods in interpretability, stability, and robustness. This work contributes to AI alignment by enhancing transparency, controllability, and trustworthiness in AI systems. Our code is available at: https://github.com/alehdaghi/PCMNet.
Authors: Shouwei Ruan, Zhenyu Wu, Yao Huang, Ruochen Zhang, Yitong Sun, Caixin Kang, Shiji Zhao, Xingxing Wei
Abstract: Content safety is a fundamental challenge for text-to-image (T2I) models, yet prevailing methods enforce a debilitating trade-off between safety and generation quality. We argue that mitigating this trade-off hinges on addressing systemic challenges in current T2I safety alignment across data, methods, and evaluation protocols. To this end, we introduce a unified framework for synergistic safety alignment. First, to overcome the flawed data paradigm that provides biased optimization signals, we develop LibraAlign-100K, the first large-scale dataset with dual annotations for safety and quality. Second, to address the myopic optimization of existing methods focus solely on safety reward, we propose Synergistic Preference Optimization (T2I-SPO), a novel alignment algorithm that extends the DPO paradigm with a composite reward function that integrates generation safety and quality to holistically model user preferences. Finally, to overcome the limitations of quality-agnostic and binary evaluation in current protocols, we introduce the Unified Alignment Score, a holistic, fine-grained metric that fairly quantifies the balance between safety and generative capability. Extensive experiments demonstrate that T2I-SPO achieves state-of-the-art safety alignment against a wide range of NSFW concepts, while better maintaining the model's generation quality and general capability
Authors: Bowei Zhang, Lei Ke, Adam W. Harley, Katerina Fragkiadaki
Abstract: We introduce TAPIP3D, a novel approach for long-term 3D point tracking in monocular RGB and RGB-D videos. TAPIP3D represents videos as camera-stabilized spatio-temporal feature clouds, leveraging depth and camera motion information to lift 2D video features into a 3D world space where camera movement is effectively canceled out. Within this stabilized 3D representation, TAPIP3D iteratively refines multi-frame motion estimates, enabling robust point tracking over long time horizons. To handle the irregular structure of 3D point distributions, we propose a 3D Neighborhood-to-Neighborhood (N2N) attention mechanism - a 3D-aware contextualization strategy that builds informative, spatially coherent feature neighborhoods to support precise trajectory estimation. Our 3D-centric formulation significantly improves performance over existing 3D point tracking methods and even surpasses state-of-the-art 2D pixel trackers in accuracy when reliable depth is available. The model supports inference in both camera-centric (unstabilized) and world-centric (stabilized) coordinates, with experiments showing that compensating for camera motion leads to substantial gains in tracking robustness. By replacing the conventional 2D square correlation windows used in prior 2D and 3D trackers with a spatially grounded 3D attention mechanism, TAPIP3D achieves strong and consistent results across multiple 3D point tracking benchmarks. Project Page: https://tapip3d.github.io
Authors: Mingxia Zhan, Li Zhang, Yingjie Wang, Xiaomeng Chu, Beibei Wang, Yanyong Zhang
Abstract: Monocular depth estimation (MDE) aims to infer per-pixel depth from a single RGB image. While diffusion models have advanced MDE with impressive generalization, they often exhibit limitations in accurately reconstructing far-range regions. This difficulty arises from two key challenges. First, the implicit multi-scale processing in standard spatial-domain models can be insufficient for preserving the fine-grained, high-frequency details crucial for distant structures. Second, the intrinsic long-tail distribution of depth data imposes a strong training bias towards more prevalent near-range regions. To address these, we propose VistaDepth, a novel diffusion framework designed for balanced and accurate depth perception. We introduce two key innovations. First, the Latent Frequency Modulation (LFM) module enhances the model's ability to represent high-frequency details. It operates by having a lightweight network predict a dynamic, content-aware spectral filter to refine latent features, thereby improving the reconstruction of distant structures. Second, our BiasMap mechanism introduces an adaptive reweighting of the diffusion loss strategically scaled across diffusion timesteps. It further aligns the supervision with the progressive denoising process, establishing a more consistent learning signal. As a result, it mitigates data bias without sacrificing training stability. Experiments show that VistaDepth achieves state-of-the-art performance for diffusion-based MDE, particularly excelling in reconstructing detailed and accurate depth in far-range regions.
Authors: Colton R. Crum, Adam Czajka
Abstract: Despite several algorithmic advances in the training of convolutional neural networks (CNNs) over the years, their generalization capabilities are still subpar across several pertinent domains, particularly within open-set tasks often found in biometric and medical contexts. On the contrary, humans have an uncanny ability to generalize to unknown visual stimuli. The efficient coding hypothesis posits that early visual structures (retina, Lateral Geniculate Nucleus, and primary visual cortex) transform inputs to reduce redundancy and maximize information efficiency. This mechanism of redundancy minimization in early vision was the inspiration for CNN regularization techniques that force convolutional kernels to be orthogonal. However, the existing works rely upon matrix projections, architectural modifications, or specific weight initializations, which frequently overtly constrain the network's learning process and excessively increase the computational load during loss function calculation. In this paper, we introduce a flexible and lightweight approach that regularizes a subset of first-layer convolutional filters by making them pairwise-orthogonal, which reduces the redundancy of the extracted features but at the same time prevents putting excessive constraints on the network. We evaluate the proposed method on three open-set visual tasks (anomaly detection in chest X-ray images, synthetic face detection, and iris presentation attack detection) and observe an increase in the generalization capabilities of models trained with the proposed regularizer compared to state-of-the-art kernel orthogonalization approaches. We offer source codes along with the paper.
Authors: Yuming Zhao, Qijian Zhang, Junhui Hou, Jiazhi Xia, Wenping Wang, Ying He
Abstract: Surface parameterization is a fundamental geometry processing task, laying the foundations for the visual presentation of 3D assets and numerous downstream shape analysis scenarios. Conventional parameterization approaches demand high-quality mesh triangulation and are restricted to certain simple topologies unless additional surface cutting and decomposition are provided. In practice, the optimal configurations (e.g., type of parameterization domains, distribution of cutting seams, number of mapping charts) may vary drastically with different surface structures and task characteristics, thus requiring more flexible and controllable processing pipelines. To this end, this paper introduces FlexPara, an unsupervised neural optimization framework to achieve both global and multi-chart surface parameterizations by establishing point-wise mappings between 3D surface points and adaptively-deformed 2D UV coordinates. We ingeniously design and combine a series of geometrically-interpretable sub-networks, with specific functionalities of cutting, deforming, unwrapping, and wrapping, to construct a bi-directional cycle mapping framework for global parameterization without the need for manually specified cutting seams. Furthermore, we construct a multi-chart parameterization framework with adaptively-learned chart assignment. Extensive experiments demonstrate the universality, superiority, and inspiring potential of our neural surface parameterization paradigm. The code will be publicly available at https://github.com/AidenZhao/FlexPara
Authors: Rayson Laroca, Marcelo dos Santos, David Menotti
Abstract: Detecting small drones, often indistinguishable from birds, is crucial for modern surveillance. This work introduces a drone detection methodology built upon the medium-sized YOLOv11 object detection model. To enhance its performance on small targets, we implemented a multi-scale approach in which the input image is processed both as a whole and in segmented parts, with subsequent prediction aggregation. We also utilized a copy-paste data augmentation technique to enrich the training dataset with diverse drone and bird examples. Finally, we implemented a post-processing technique that leverages frame-to-frame consistency to mitigate missed detections. The proposed approach attained first place in the 8th WOSDETC Drone-vs-Bird Detection Grand Challenge, held at the 2025 International Joint Conference on Neural Networks (IJCNN), showcasing its capability to detect drones in complex environments effectively.
Authors: Junlong Ren, Gangjian Zhang, Yu Hu, Jian Shu, Hui Xiong, Hao Wang
Abstract: Partially Relevant Video Retrieval (PRVR) aims to retrieve the target video that is partially relevant to the text query. The primary challenge in PRVR arises from the semantic asymmetry between textual and visual modalities, as videos often contain substantial content irrelevant to the query. Existing methods coarsely align paired videos and text queries to construct the semantic space, neglecting the critical cross-modal dual nature inherent in this task: inter-sample correlation and intra-sample redundancy. To this end, we propose a novel PRVR framework to systematically exploit these two characteristics. Our framework consists of three core modules. First, the Inter Correlation Enhancement (ICE) module captures inter-sample correlation by identifying semantically similar yet unpaired text queries and video moments, combining them to form pseudo-positive pairs for more robust semantic space construction. Second, the Intra Redundancy Mining (IRM) module mitigates intra-sample redundancy by mining redundant moment features and distinguishing them from query-relevant moments, encouraging the model to learn more discriminative representations. Finally, to reinforce these modules, we introduce the Temporal Coherence Prediction (TCP) module, which enhances discrimination of fine-grained moment-level semantics by training the model to predict the original temporal order of randomly shuffled video frames and moments. Extensive experiments demonstrate the superiority of our method, achieving state-of-the-art results.
Authors: Shijun Zhou, Xing Xie, Baojie Fan, Jiandong Tian
Abstract: Due to the domain gap between real-world and synthetic hazy images, current data-driven dehazing algorithms trained on synthetic datasets perform well on synthetic data but struggle to generalize to real-world scenarios. To address this challenge, we propose \textbf{I}mage \textbf{D}ehazing \textbf{D}iffusion \textbf{M}odels (IDDM), a novel diffusion process that incorporates the atmospheric scattering model into noise diffusion. IDDM aims to use the gradual haze formation process to help the denoising Unet robustly learn the distribution of clear images from the conditional input hazy images. We design a specialized training strategy centered around IDDM. Diffusion models are leveraged to bridge the domain gap from synthetic to real-world, while the atmospheric scattering model provides physical guidance for haze formation. During the forward process, IDDM simultaneously introduces haze and noise into clear images, and then robustly separates them during the sampling process. By training with physics-guided information, IDDM shows the ability of domain generalization, and effectively restores the real-world hazy images despite being trained on synthetic datasets. Extensive experiments demonstrate the effectiveness of our method through both quantitative and qualitative comparisons with state-of-the-art approaches.
Authors: Xinzhu Ma, Cheng Wang, Chen Tang, Bin Wang, Shixiang Tang, Yuan Meng, Yunhong Wang, Di Huang
Abstract: Recovering CAD models from point clouds requires reconstructing their topology and sketch-based extrusion primitives. A dominant paradigm for representing sketches involves implicit neural representations such as Signed Distance Fields (SDFs). However, this indirect approach inherently struggles with precision, leading to unintended curved edges and models that are difficult to edit. In this paper, we propose Point2Primitive, a framework that learns to directly predict the explicit, parametric primitives of CAD models. Our method treats sketch reconstruction as a set prediction problem, employing a improved transformer-based decoder with explicit position queries to directly detect and predict the fundamental sketch curves (i.e., type and parameter) from the point cloud. Instead of approximating a continuous field, we formulate curve parameters as explicit position queries, which are optimized autoregressively to achieve high accuracy. The overall topology is rebuilt via extrusion segmentation. Extensive experiments demonstrate that this direct prediction paradigm significantly outperforms implicit methods in both primitive accuracy and overall geometric fidelity.
Authors: Wenqi Wang, Reuben Tan, Pengyue Zhu, Jianwei Yang, Zhengyuan Yang, Lijuan Wang, Andrey Kolobov, Jianfeng Gao, Boqing Gong
Abstract: Spatial intelligence (SI) represents a cognitive ability encompassing the visualization, manipulation, and reasoning about spatial relationships, underpinning disciplines from neuroscience to robotics. We introduce SITE, a benchmark dataset towards SI Thorough Evaluation in a standardized format of multi-choice visual question-answering, designed to assess large vision-language models' spatial intelligence across diverse visual modalities (single-image, multi-image, and video) and SI factors (figural to environmental scales, spatial visualization and orientation, intrinsic and extrinsic, static and dynamic). Our approach to curating the benchmark combines a bottom-up survey about 31 existing datasets and a top-down strategy drawing upon three classification systems in cognitive science, which prompt us to design two novel types of tasks about view-taking and dynamic scenes. Extensive experiments reveal that leading models fall behind human experts especially in spatial orientation, a fundamental SI factor. Moreover, we demonstrate a positive correlation between a model's spatial reasoning proficiency and its performance on an embodied AI task.
Authors: Hongyang Wang, Yichen Shi, Zhuofu Tao, Yuhao Gao, Liepiao Zhang, Xun Lin, Jun Feng, Xiaochen Yuan, Zitong Yu, Xiaochun Cao
Abstract: Face anti-spoofing (FAS) is crucial for protecting facial recognition systems from presentation attacks. Previous methods approached this task as a classification problem, lacking interpretability and reasoning behind the predicted results. Recently, multimodal large language models (MLLMs) have shown strong capabilities in perception, reasoning, and decision-making in visual tasks. However, there is currently no universal and comprehensive MLLM and dataset specifically designed for FAS task. To address this gap, we propose FaceShield, a MLLM for FAS, along with the corresponding pre-training and supervised fine-tuning (SFT) datasets, FaceShield-pre10K and FaceShield-sft45K. FaceShield is capable of determining the authenticity of faces, identifying types of spoofing attacks, providing reasoning for its judgments, and detecting attack areas. Specifically, we employ spoof-aware vision perception (SAVP) that incorporates both the original image and auxiliary information based on prior knowledge. We then use an prompt-guided vision token masking (PVTM) strategy to random mask vision tokens, thereby improving the model's generalization ability. We conducted extensive experiments on three benchmark datasets, demonstrating that FaceShield significantly outperforms previous deep learning models and general MLLMs on four FAS tasks, i.e., coarse-grained classification, fine-grained classification, reasoning, and attack localization. Our instruction datasets, protocols, and codes will be released at https://github.com/Why0912/FaceShield.
Authors: Fu-Yun Wang, Keqiang Sun, Yao Teng, Xihui Liu, Jiale Yuan, Jiaming Song, Hongsheng Li
Abstract: Diffusion models have demonstrated remarkable success in various visual generation tasks, including image, video, and 3D content generation. Preference optimization (PO) is a prominent and growing area of research that aims to align these models with human preferences. While existing PO methods primarily concentrate on producing favorable outputs, they often overlook the significance of classifier-free guidance (CFG) in mitigating undesirable results. Diffusion-NPO addresses this gap by introducing negative preference optimization (NPO), training models to generate outputs opposite to human preferences and thereby steering them away from unfavorable outcomes through CFG. However, prior NPO approaches rely on costly and fragile procedures for obtaining explicit preference annotations (e.g., manual pairwise labeling or reward model training), limiting their practicality in domains where such data are scarce or difficult to acquire. In this work, we propose Self-NPO, specifically truncated diffusion fine-tuning, a data-free approach of negative preference optimization by directly learning from the model itself, eliminating the need for manual data labeling or reward model training. This data-free approach is highly efficient (less than 1% training cost of Diffusion-NPO) and achieves comparable performance to Diffusion-NPO in a data-free manner. We demonstrate that Self-NPO integrates seamlessly into widely used diffusion models, including SD1.5, SDXL, and CogVideoX, as well as models already optimized for human preferences, consistently enhancing both their generation quality and alignment with human preferences. Code is available at https://github.com/G-U-N/Diffusion-NPO.
Authors: Ke Li, Di Wang, Xiaowei Wang, Zhihong Wu, Yiming Zhang, Yifeng Wang, Quan Wang
Abstract: Drone-view geo-localization (DVGL) aims to match images of the same geographic location captured from drone and satellite perspectives. Despite recent advances, DVGL remains challenging due to significant appearance changes and spatial distortions caused by viewpoint variations. Existing methods typically assume that drone and satellite images can be directly aligned in a shared feature space via contrastive learning. Nonetheless, this assumption overlooks the inherent conflicts induced by viewpoint discrepancies, resulting in extracted features containing inconsistent information that hinders precise localization. In this study, we take a manifold learning perspective and model $\textit{the feature space of cross-view images as a composite manifold jointly governed by content and viewpoint}$. Building upon this insight, we propose $\textbf{CVD}$, a new DVGL framework that explicitly disentangles $\textit{content}$ and $\textit{viewpoint}$ factors. To promote effective disentanglement, we introduce two constraints: $\textit{(i)}$ an intra-view independence constraint that encourages statistical independence between the two factors by minimizing their mutual information; and $\textit{(ii)}$ an inter-view reconstruction constraint that reconstructs each view by cross-combining $\textit{content}$ and $\textit{viewpoint}$ from paired images, ensuring factor-specific semantics are preserved. As a plug-and-play module, CVD integrates seamlessly into existing DVGL pipelines and reduces inference latency. Extensive experiments on University-1652 and SUES-200 show that CVD exhibits strong robustness and generalization across various scenarios, viewpoints and altitudes, with further evaluations on CVUSA and CVACT confirming consistent improvements.
Authors: Bo Yang, Hengwei Zhang, Jindong Wang, Yuchen Ren, Chenhao Lin, Chao Shen, Zhengyu Zhao
Abstract: In surrogate ensemble attacks, using more surrogate models yields higher transferability but lower resource efficiency. This practical trade-off between transferability and efficiency has largely limited existing attacks despite many pre-trained models are easily accessible online. In this paper, we argue that such a trade-off is caused by an unnecessary common assumption, i.e., all models should be \textit{identical} across iterations. By lifting this assumption, we can use as many surrogates as we want to unleash transferability without sacrificing efficiency. Concretely, we propose Selective Ensemble Attack (SEA), which dynamically selects diverse models (from easily accessible pre-trained models) across iterations based on our new interpretation of decoupling within-iteration and cross-iteration model diversity. In this way, the number of within-iteration models is fixed for maintaining efficiency, while only cross-iteration model diversity is increased for higher transferability. Experiments on ImageNet demonstrate the superiority of SEA in various scenarios. For example, when dynamically selecting 4 from 20 accessible models, SEA yields 8.5% higher transferability than existing attacks under the same efficiency. The superiority of SEA also generalizes to real-world systems, such as commercial vision APIs and large vision-language models. Overall, SEA opens up the possibility of adaptively balancing transferability and efficiency according to specific resource requirements.
Authors: Pengcheng Pan, Yonekura Shogo, Yasuo Kuniyoshi
Abstract: Inspired by foveal vision, hard attention models promise interpretability and parameter economy. However, existing models like the Recurrent Model of Visual Attention (RAM) and Deep Recurrent Attention Model (DRAM) failed to model the hierarchy of human vision system, that compromise on the visual exploration dynamics. As a result, they tend to produce attention that are either overly fixational or excessively saccadic, diverging from human eye movement behavior. In this paper, we propose a Multi-Level Recurrent Attention Model (MRAM), a novel hard attention framework that explicitly models the neural hierarchy of human visual processing. By decoupling the function of glimpse location generation and task execution in two recurrent layers, MRAM emergent a balanced behavior between fixation and saccadic movement. Our results show that MRAM not only achieves more human-like attention dynamics, but also consistently outperforms CNN, RAM and DRAM baselines on standard image classification benchmarks.
Authors: Chaoyang Wang, Xiangtai Li, Lu Qi, Xiaofan Lin, Jinbin Bai, Qianyu Zhou, Yunhai Tong
Abstract: Recent progress in panoramic image generation has underscored two critical limitations in existing approaches. First, most methods are built upon diffusion models, which are inherently ill-suited for equirectangular projection (ERP) panoramas due to the violation of the identically and independently distributed (i.i.d.) Gaussian noise assumption caused by their spherical mapping. Second, these methods often treat text-conditioned generation (text-to-panorama) and image-conditioned generation (panorama outpainting) as separate tasks, relying on distinct architectures and task-specific data. In this work, we propose a unified framework, Panoramic AutoRegressive model (PAR), which leverages masked autoregressive modeling to address these challenges. PAR avoids the i.i.d. assumption constraint and integrates text and image conditioning into a cohesive architecture, enabling seamless generation across tasks. To address the inherent discontinuity in existing generative models, we introduce circular padding to enhance spatial coherence and propose a consistency alignment strategy to improve generation quality. Extensive experiments demonstrate competitive performance in text-to-image generation and panorama outpainting tasks while showcasing promising scalability and generalization capabilities.
Authors: Hancong Jin, Zihan Cao, Liang-jian Deng, Jingjing Li
Abstract: Pansharpening seeks to fuse high-resolution panchromatic (PAN) and low-resolution multispectral (LRMS) images into a single image with both fine spatial and rich spectral detail. Despite progress in deep learning-based approaches, existing methods often fail to capture global priors inherent in remote sensing data distributions. Diffusion-based models have recently emerged as promising solutions due to their powerful distribution mapping capabilities, however, they suffer from heavy inference latency. We introduce KSDiff, a fast kernel-space diffusion framework that generates convolutional kernels enriched with global context to enhance pansharpening quality and accelerate inference. Specifically, KSDiff constructs these kernels through the integration of a low-rank core tensor generator and a unified factor generator, orchestrated by a structure-aware multi-head attention mechanism. We further introduce a two-stage training strategy tailored for pansharpening, facilitating integration into existing pansharpening architectures. Experiments show that KSDiff achieves superior performance compared to recent promising methods, and with over $500 \times$ faster inference than diffusion-based pansharpening baselines. Ablation studies, visualizations and further evaluations substantiate the effectiveness of our approach. Code will be released upon possible acceptance.
Authors: Ta\"iga Gon\c{c}alves, Tomo Miyazaki, Shinichiro Omachi
Abstract: Multi-targeted adversarial attacks aim to mislead classifiers toward specific target classes using a single perturbation generator with a conditional input specifying the desired target class. Existing methods face two key limitations: (1) a single generator supports only a limited number of predefined target classes, and (2) it requires access to the victim model's training data to learn target class semantics. This dependency raises data leakage concerns in practical black-box scenarios where the training data is typically private. To address these limitations, we propose a novel Cross-Domain Multi-Targeted Attack (CD-MTA) that can generate perturbations toward arbitrary target classes, even those that do not exist in the attacker's training data. CD-MTA is trained on a single public dataset but can perform targeted attacks on black-box models trained on different datasets with disjoint and unknown class sets. Our method requires only a single example image that visually represents the desired target class, without relying its label, class distribution or pretrained embeddings. We achieve this through a Feature Injection Module (FIM) and class-agnostic objectives which guide the generator to extract transferable, fine-grained features from the target image without inferring class semantics. Experiments on ImageNet and seven additional datasets show that CD-MTA outperforms existing multi-targeted attack methods on unseen target classes in black-box and cross-domain scenarios. The code is available at https://github.com/tgoncalv/CD-MTA.
Authors: Yinjie Chen, Zipeng Yan, Chong Zhou, Bo Dai, Andrew F. Luo
Abstract: Vision Transformers (ViTs) have emerged as the dominant architecture for visual processing tasks, demonstrating excellent scalability with increased training data and model size. However, recent work has identified the emergence of artifact tokens in ViTs that are incongruous with local semantics. These anomalous tokens degrade ViT performance in tasks that require fine-grained localization or structural coherence. An effective mitigation of this issue is the addition of register tokens to ViTs, which implicitly "absorb" the artifact term during training.Given the availability of existing large-scale pre-trained ViTs, in this paper we seek add register tokens to existing models without needing to re-train from scratch, which is infeasible considering their size. Specifically, we propose Post Hoc Registers (PH-Reg), an efficient self-distillation method that integrates registers into an existing ViT without requiring additional labeled data and full retraining. PH-Reg initializes both teacher and student networks from the same pre-trained ViT. The teacher remains frozen and unmodified, while the student is augmented with randomly initialized register tokens. By applying test-time augmentation to the teacher's inputs, we generate denoised dense embeddings free of artifacts, which are then used to optimize only a small subset of unlocked student weights. We show that our approach can effectively reduce the number of artifact tokens, improving the segmentation and depth prediction of the student ViT under zero-shot and linear probing.
Authors: Claudia Cuttano, Gabriele Trivigno, Giuseppe Averta, Carlo Masone
Abstract: Few-shot segmentation aims to segment unseen object categories from just a handful of annotated examples. This requires mechanisms that can both identify semantically related objects across images and accurately produce segmentation masks. We note that Segment Anything 2 (SAM2), with its prompt-and-propagate mechanism, offers both strong segmentation capabilities and a built-in feature matching process. However, we show that its representations are entangled with task-specific cues optimized for object tracking, which impairs its use for tasks requiring higher level semantic understanding. Our key insight is that, despite its class-agnostic pretraining, SAM2 already encodes rich semantic structure in its features. We propose SANSA (Semantically AligNed Segment Anything 2), a framework that makes this latent structure explicit, and repurposes SAM2 for few-shot segmentation through minimal task-specific modifications. SANSA achieves state-of-the-art performance on few-shot segmentation benchmarks specifically designed to assess generalization, outperforms generalist methods in the popular in-context setting, supports various prompts flexible interaction via points, boxes, or scribbles, and remains significantly faster and more compact than prior approaches. Code is available at https://github.com/ClaudiaCuttano/SANSA.
Authors: Nikhil Behari, Aaron Young, Tzofi Klinghoffer, Akshat Dave, Ramesh Raskar
Abstract: Imaging system design is a complex, time-consuming, and largely manual process; LiDAR design, ubiquitous in mobile devices, autonomous vehicles, and aerial imaging platforms, adds further complexity through unique spatial and temporal sampling requirements. In this work, we propose a framework for automated, task-driven LiDAR system design under arbitrary constraints. To achieve this, we represent LiDAR configurations in a continuous six-dimensional design space and learn task-specific implicit densities in this space via flow-based generative modeling. We then synthesize new LiDAR systems by modeling sensors as parametric distributions in 6D space and fitting these distributions to our learned implicit density using expectation-maximization, enabling efficient, constraint-aware LiDAR system design. We validate our method on diverse tasks in 3D vision, enabling automated LiDAR system design across real-world-inspired applications in face scanning, robotic tracking, and object detection.
Authors: Weijie Wang, Donny Y. Chen, Zeyu Zhang, Duochao Shi, Akide Liu, Bohan Zhuang
Abstract: Feed-forward 3D Gaussian Splatting (3DGS) models have recently emerged as a promising solution for novel view synthesis, enabling one-pass inference without the need for per-scene 3DGS optimization. However, their scalability is fundamentally constrained by the limited capacity of their models, leading to degraded performance or excessive memory consumption as the number of input views increases. In this work, we analyze feed-forward 3DGS frameworks through the lens of the Information Bottleneck principle and introduce ZPressor, a lightweight architecture-agnostic module that enables efficient compression of multi-view inputs into a compact latent state $Z$ that retains essential scene information while discarding redundancy. Concretely, ZPressor enables existing feed-forward 3DGS models to scale to over 100 input views at 480P resolution on an 80GB GPU, by partitioning the views into anchor and support sets and using cross attention to compress the information from the support views into anchor views, forming the compressed latent state $Z$. We show that integrating ZPressor into several state-of-the-art feed-forward 3DGS models consistently improves performance under moderate input views and enhances robustness under dense view settings on two large-scale benchmarks DL3DV-10K and RealEstate10K. The video results, code and trained models are available on our project page: https://lhmd.top/zpressor.
Authors: Yu Huang, Junhao Chen, Shuliang Liu, Hanqian Li, Jungang Li, Qi Zheng, Aiwei Liu, Yi R. Fung, Xuming Hu
Abstract: The rapid development of Artificial Intelligence Generated Content (AIGC) has led to significant progress in video generation, but also raises serious concerns about intellectual property protection and reliable content tracing. Watermarking is a widely adopted solution to this issue, yet existing methods for video generation mainly follow a post-generation paradigm, which often fails to effectively balance the trade-off between video quality and watermark extraction. Meanwhile, current in-generation methods that embed the watermark into the initial Gaussian noise usually incur substantial additional computation. To address these issues, we propose \textbf{Video Signature} (\textsc{VidSig}), an implicit watermarking method for video diffusion models that enables imperceptible and adaptive watermark integration during video generation with almost no extra latency. Specifically, we partially fine-tune the latent decoder, where \textbf{Perturbation-Aware Suppression} (PAS) pre-identifies and freezes perceptually sensitive layers to preserve visual quality. Beyond spatial fidelity, we further enhance temporal consistency by introducing a lightweight \textbf{Temporal Alignment} module that guides the decoder to generate coherent frame sequences during fine-tuning. Experimental results show that \textsc{VidSig} achieves the best trade-off among watermark extraction accuracy, video quality, and watermark latency. It also demonstrates strong robustness against both spatial and temporal tamper, and remains stable across different video lengths and resolutions, highlighting its practicality in real-world scenarios.
Authors: Xinran Nicole Han, Ko Nishino, Todd Zickler
Abstract: Perceiving the shape and material of an object from a single image is inherently ambiguous, especially when lighting is unknown and unconstrained. Despite this, humans can often disentangle shape and material, and when they are uncertain, they often move their head slightly or rotate the object to help resolve the ambiguities. Inspired by this behavior, we introduce a novel conditional denoising-diffusion model that generates samples of shape-and-material maps from a short video of an object undergoing differential motions. Our parameter-efficient architecture allows training directly in pixel-space, and it generates many disentangled attributes of an object simultaneously. Trained on a modest number of synthetic object-motion videos with supervision on shape and material, the model exhibits compelling emergent behavior: For static observations, it produces diverse, multimodal predictions of plausible shape-and-material maps that capture the inherent ambiguities; and when objects move, the distributions converge to more accurate explanations. The model also produces high-quality shape-and-material estimates for less ambiguous, real-world objects. By moving beyond single-view to continuous motion observations, and by using generative perception to capture visual ambiguities, our work suggests ways to improve visual reasoning in physically-embodied systems.
Authors: Sining Chen, Yilei Shi, Xiao Xiang Zhu
Abstract: Monocular height estimation provides an efficient and cost-effective solution for three-dimensional perception in remote sensing. However, training deep neural networks for this task demands abundant annotated data, while high-quality labels are scarce and typically available only in developed regions, which limits model generalization and constrains their applicability at large scales. This work addresses the problem by leveraging imperfect labels from out-of-domain regions to train pixel-wise height estimation networks, which may be incomplete, inexact, or inaccurate compared to high-quality annotations. We introduce an ensemble-based pipeline compatible with any monocular height estimation network, featuring architecture and loss functions specifically designed to leverage information in noisy labels through weak supervision, utilizing balanced soft losses and ordinal constraints. Experiments on two datasets -- DFC23 (0.5--1 m) and GBH (3 m) -- show that our method achieves more consistent cross-domain performance, reducing average RMSE by up to 22.94% on DFC23 and 18.62% on GBH compared with baselines. Ablation studies confirm the contribution of each design component.
Authors: Shuhan Xu, Siyuan Liang, Hongling Zheng, Aishan Liu, Xinbiao Wang, Yong Luo, Fu Lin, Leszek Rutkowski, Dacheng Tao
Abstract: Visual language models (VLMs) have made significant progress in image captioning tasks, yet recent studies have found they are vulnerable to backdoor attacks. Attackers can inject undetectable perturbations into the data during inference, triggering abnormal behavior and generating malicious captions. These attacks are particularly challenging to detect and defend against due to the stealthiness and cross-modal propagation of the trigger signals. In this paper, we identify two key vulnerabilities by analyzing existing attack patterns: (1) the model exhibits abnormal attention concentration on certain regions of the input image, and (2) backdoor attacks often induce semantic drift and sentence incoherence. Based on these insights, we propose Semantic Reward Defense (SRD), a reinforcement learning framework that mitigates backdoor behavior without requiring any prior knowledge of trigger patterns. SRD learns to apply discrete perturbations to sensitive contextual regions of image inputs via a deep Q-network policy, aiming to confuse attention and disrupt the activation of malicious paths. To guide policy optimization, we design a reward signal named semantic fidelity score, which jointly assesses the semantic consistency and linguistic fluency of the generated captions, encouraging the agent to achieve a robust yet faithful output. SRD offers a trigger-agnostic, policy-interpretable defense paradigm that effectively mitigates local (TrojVLM) and global (Shadowcast) backdoor attacks, reducing ASR to 3.6% and 5.6% respectively, with less than 15% average CIDEr drop on the clean inputs. Our codes can be found at https://github.com/Ciconey/SRD.git.
Authors: Hong Gao, Yiming Bao, Xuezhen Tu, Bin Zhong, Linan Yue, Minling Zhang
Abstract: Current multimodal large language models (MLLMs) struggle with hour-level video understanding, facing significant challenges not only in modeling the substantial information volume of long videos but also in overcoming the memory wall and resource constraints during both training and inference. Although recent training-free approaches have alleviated resource demands by compressing visual features, their reliance on incomplete visual information limits the performance potential. To address these limitations, we propose Adaptive Pivot Visual information Retrieval (APVR), a training-free framework that hierarchically retrieves and retains sufficient and important visual information. It breakthroughs the memory wall limitation via two complementary components: Pivot Frame Retrieval employs query expansion and iterative spatio-semantic confidence scoring to identify relevant video frames, and Pivot Token Retrieval performs query-aware attention-driven token selection within up to 1024 pivot frames. This dual granularity approach enables the processing of hour-long videos while maintaining semantic fidelity. Experimental validations on three different baseline MLLMs demonstrate significant performance improvements up to 9.5\%, 4.6\% and 9.7\% on LongVideoBench, VideoMME and MLVU, respectively. APVR achieves state-of-the-art results for both training-free and training-based approaches.
Authors: Seonho Lee, Jiho Choi, Inha Kang, Jiwook Kim, Junsung Park, Hyunjung Shim
Abstract: Vision-Language Models (VLMs) have shown remarkable performance on diverse visual and linguistic tasks, yet they remain fundamentally limited in their understanding of 3D spatial structures. We propose Geometric Distillation, a lightweight, annotation-free fine-tuning framework that injects human-inspired geometric cues into pretrained VLMs without modifying their architecture. By distilling (1) sparse correspondences, (2) relative depth relations, and (3) dense cost volumes from off-the-shelf 3D foundation models (e.g., MASt3R, VGGT), our method shapes representations to be geometry-aware while remaining compatible with natural image-text inputs. Through extensive evaluations on 3D vision-language reasoning and 3D perception benchmarks, our method consistently outperforms prior approaches, achieving improved 3D spatial reasoning with significantly lower computational cost. Our work demonstrates a scalable and efficient path to bridge 2D-trained VLMs with 3D understanding, opening up wider use in spatially grounded multimodal tasks.
Authors: Paul Couairon, Loick Chambon, Louis Serrano, Jean-Emmanuel Haugeard, Matthieu Cord, Nicolas Thome
Abstract: Foundation Vision Encoders have become essential for a wide range of dense vision tasks. However, their low-resolution spatial feature outputs necessitate feature upsampling to produce the high-resolution modalities required for downstream tasks. In this work, we introduce JAFAR, a lightweight and flexible feature upsampler that enhances the spatial resolution of visual features from any Foundation Vision Encoder to an arbitrary target resolution. JAFAR employs an attention-based module designed to promote semantic alignment between high-resolution queries, derived from low-level image features, and semantically enriched low-resolution keys, using Spatial Feature Transform (SFT) modulation. Notably, despite the absence of high-resolution supervision, we demonstrate that learning at low upsampling ratios and resolutions generalizes remarkably well to significantly higher output scales. Extensive experiments show that JAFAR effectively recovers fine-grained spatial details and consistently outperforms existing feature upsampling methods across a diverse set of downstream tasks. Project page at https://jafar-upsampler.github.io
Authors: Chade Li, Pengju Zhang, Yihong Wu
Abstract: The rapid advancement of vision-language models (VLMs) in 3D domains has accelerated research in text-query-guided point cloud processing, though existing methods underperform in point-level segmentation due to inadequate 3D-text alignment that limits local feature-text context linking. To address this limitation, we propose MR-COSMO, a Visual-Text Memory Recall and Direct CrOSs-MOdal Alignment Method for Query-Driven 3D Segmentation, establishing explicit alignment between 3D point clouds and text/2D image data through a dedicated direct cross-modal alignment module while implementing a visual-text memory module with specialized feature banks. This direct alignment mechanism enables precise fusion of geometric and semantic features, while the memory module employs specialized banks storing text features, visual features, and their correspondence mappings to dynamically enhance scene-specific representations via attention-based knowledge recall. Comprehensive experiments across 3D instruction, reference, and semantic segmentation benchmarks confirm state-of-the-art performance.
Authors: Feng Lin, Marco Chen, Haokui Zhang, Xiaotian Yu, Guangming Lu, Rong Xiao
Abstract: This paper investigates the role of attention heads in CLIP's image encoder. Building on interpretability studies, we conduct an exhaustive analysis and find that certain heads, distributed across layers, are detrimental to the resulting representations. To mitigate their impact, we propose a simple yet effective Attention Ablation Technique (AAT) that suppresses selected heads by directly manipulating their attention weights. By incorporating two complementary strategies tailored to different application scenarios, AAT enables the systematic identification and ablation of harmful heads with minimal overhead. Experiments show that AAT consistently improves downstream performance across diverse domains, boosting recall by up to 11.1% on cross-modal retrieval benchmarks. These results highlight that AAT can effectively refine large-scale VLMs with virtually no extra inference cost, while yielding semantically meaningful patterns that align with existing interpretability findings.
Authors: Colin Laganier, Liam Fletcher, Elim Kwan, Richard Walters, Victoria Nockles
Abstract: Rapid analysis of satellite imagery within minutes-to-hours of acquisition is increasingly vital for many remote sensing applications, and is an essential component for developing next-generation autonomous and distributed satellite systems. On-satellite machine learning (ML) has the potential for such rapid analysis, by overcoming latency associated with intermittent satellite connectivity to ground stations or relay satellites, but state-of-the-art models are often too large or power-hungry for on-board deployment. Vessel detection using Synthetic Aperture Radar (SAR) is a critical time-sensitive application in maritime security that exemplifies this challenge. SAR vessel detection has previously been demonstrated only by ML models that either are too large for satellite deployment, have not been developed for sufficiently low-power hardware, or have only been tested on small SAR datasets that do not sufficiently represent the difficulty of the real-world task. Here we systematically explore a suite of architectural adaptations to develop a novel YOLOv8 architecture optimized for this task and FPGA-based processing. We deploy our model on a Kria KV260 MPSoC, and show it can analyze a ~700 megapixel SAR image in less than a minute, within common satellite power constraints (<10W). Our model has detection and classification performance only ~2% and 3% lower than values from state-of-the-art GPU-based models on the largest and most diverse open SAR vessel dataset, xView3-SAR, despite being ~50 and ~2500 times more computationally efficient. This work represents a key contribution towards on-satellite ML for time-critical SAR analysis, and more autonomous, scalable satellites.
Authors: Zach Eidex, Mojtaba Safari, Tonghe Wang, Vanessa Wildman, David S. Yu, Hui Mao, Erik Middlebrooks, Aparna Kesarwala, Xiaofeng Yang
Abstract: Purpose: Ultra-high-field 7T MRI offers improved resolution and contrast over standard clinical field strengths (1.5T, 3T). However, 7T scanners are costly, scarce, and introduce additional challenges such as susceptibility artifacts. We propose an efficient transformer-based model (7T-Restormer) to synthesize 7T-quality T1-maps from routine 1.5T or 3T T1-weighted (T1W) images. Methods: Our model was validated on 35 1.5T and 108 3T T1w MRI paired with corresponding 7T T1 maps of patients with confirmed MS. A total of 141 patient cases (32,128 slices) were randomly divided into 105 (25; 80) training cases (19,204 slices), 19 (5; 14) validation cases (3,476 slices), and 17 (5; 14) test cases (3,145 slices) where (X; Y) denotes the patients with 1.5T and 3T T1W scans, respectively. The synthetic 7T T1 maps were compared against the ResViT and ResShift models. Results: The 7T-Restormer model achieved a PSNR of 26.0 +/- 4.6 dB, SSIM of 0.861 +/- 0.072, and NMSE of 0.019 +/- 0.011 for 1.5T inputs, and 25.9 +/- 4.9 dB, and 0.866 +/- 0.077 for 3T inputs, respectively. Using 10.5 M parameters, our model reduced NMSE by 64 % relative to 56.7M parameter ResShift (0.019 vs 0.052, p = <.001 and by 41 % relative to 70.4M parameter ResViT (0.019 vs 0.032, p = <.001) at 1.5T, with similar advantages at 3T (0.021 vs 0.060 and 0.033; p < .001). Training with a mixed 1.5 T + 3 T corpus was superior to single-field strategies. Restricting the model to 1.5T increased the 1.5T NMSE from 0.019 to 0.021 (p = 1.1E-3) while training solely on 3T resulted in lower performance on input 1.5T T1W MRI. Conclusion: We propose a novel method for predicting quantitative 7T MP2RAGE maps from 1.5T and 3T T1W scans with higher quality than existing state-of-the-art methods. Our approach makes the benefits of 7T MRI more accessible to standard clinical workflows.
Authors: Ali Hojjat, Janek Haberer, Soren Pirk, Olaf Landsiedel
Abstract: ViTs deliver SOTA performance, yet their fixed computational budget prevents scalable deployment across heterogeneous hardware. Recent Matryoshka-style Transformer architectures mitigate this by embedding nested subnetworks within a single model to enable scalable inference. However, these models allocate the same amount of compute to all inputs, regardless of their complexity, which leads to inefficiencies. To address this, we introduce ThinkingViT, a nested ViT architecture that employs progressive thinking stages to dynamically adjust inference computation based on input difficulty. ThinkingViT first activates a small subset of the most important attention heads to produce an initial prediction. If the prediction confidence exceeds a predefined threshold, inference terminates early. Otherwise, within the same backbone, it activates a larger subset of attention heads and conducts a new forward pass. This process continues iteratively until the model reaches the predefined confidence level or exhausts its maximum capacity. To boost the performance of subsequent rounds, we introduce a Token Recycling approach that fuses the input embeddings with the embeddings from the previous stage. Experiments show that ThinkingViT surpasses nested baselines by up to 2.0 percentage points (p.p.) in accuracy at the same throughput and by up to 2.9 p.p. at equal GMACs on ImageNet-1K. We show that the backbone-preserving design of ThinkingViT allows it to serve as a plug-in upgrade for ViTs in downstream tasks such as semantic segmentation. We also demonstrate that ThinkingViT transfers effectively to other architectures such as Swin. The source code is available at https://github.com/ds-kiel/ThinkingViT.
Authors: Ryan Xu, Dongyang Jin, Yancheng Bai, Rui Lan, Xu Duan, Lei Sun, Xiangxiang Chu
Abstract: Controllable image synthesis, which enables fine-grained control over generated outputs, has emerged as a key focus in visual generative modeling. However, controllable generation remains challenging for Visual Autoregressive (VAR) models due to their hierarchical, next-scale prediction style. Existing VAR-based methods often suffer from inefficient control encoding and disruptive injection mechanisms that compromise both fidelity and efficiency. In this work, we present SCALAR, a controllable generation method based on VAR, incorporating a novel Scale-wise Conditional Decoding mechanism. SCALAR leverages a pretrained image encoder to extract semantic control signal encodings, which are projected into scale-specific representations and injected into the corresponding layers of the VAR backbone. This design provides persistent and structurally aligned guidance throughout the generation process. Building on SCALAR, we develop SCALAR-Uni, a unified extension that aligns multiple control modalities into a shared latent space, supporting flexible multi-conditional guidance in a single model. Extensive experiments show that SCALAR achieves superior generation quality and control precision across various tasks. The code is released at https://github.com/AMAP-ML/SCALAR.
Authors: Ao Li, Yuxiang Duan, Jinghui Zhang, Congbo Ma, Yutong Xie, Gustavo Carneiro, Mohammad Yaqub, Hu Wang
Abstract: Large Vision-Language Models (LVLMs) have advanced multimodal learning but face high computational costs due to the large number of visual tokens, motivating token pruning to improve inference efficiency. The key challenge lies in identifying which tokens are truly important. Most existing approaches rely on attention-based criteria to estimate token importance. However, they inherently suffer from certain limitations, such as positional bias. In this work, we explore a new perspective on token importance based on token transitions in LVLMs. We observe that the transition of token representations provides a meaningful signal of semantic information. Based on this insight, we propose TransPrune, a training-free and efficient token pruning method. Specifically, TransPrune progressively prunes tokens by assessing their importance through a combination of Token Transition Variation (TTV)-which measures changes in both the magnitude and direction of token representations-and Instruction-Guided Attention (IGA), which measures how strongly the instruction attends to image tokens via attention. Extensive experiments demonstrate that TransPrune achieves comparable multimodal performance to original LVLMs, such as LLaVA-v1.5 and LLaVA-Next, across eight benchmarks, while reducing inference TFLOPs by more than half. Moreover, TTV alone can serve as an effective criterion without relying on attention, achieving performance comparable to attention-based methods. The code will be made publicly available upon acceptance of the paper at https://github.com/liaolea/TransPrune.
Authors: YiZhou Li
Abstract: We present a compact encoder for image categorization that emphasizes computation economy through content-conditioned multi-pass processing. The model employs a single lightweight core block that can be re-applied a small number of times, while a simple score-based selector decides whether further passes are beneficial for each region unit in the feature map. This design provides input-conditioned depth without introducing heavy auxiliary modules or specialized pretraining. On standard benchmarks, the approach attains competitive accuracy with reduced parameters, lower floating-point operations, and faster inference compared to similarly sized baselines. The method keeps the architecture minimal, implements module reuse to control footprint, and preserves stable training via mild regularization on selection scores. We discuss implementation choices for efficient masking, pass control, and representation caching, and show that the multi-pass strategy transfers well to several datasets without requiring task-specific customization.
Authors: Yuqi Pang, Bowen Yang, Yun Cao, Rong Fan, Xiaoyu Li, Chen He
Abstract: Vision large language models (VLLMs) are focusing primarily on handling complex and fine-grained visual information by incorporating advanced vision encoders and scaling up visual models. However, these approaches face high training and inference costs, as well as challenges in extracting visual details, effectively bridging across modalities. In this work, we propose a novel visual framework, MoCHA, to address these issues. Our framework integrates four vision backbones (i.e., CLIP, SigLIP, DINOv2 and ConvNeXt) to extract complementary visual features and is equipped with a sparse Mixture of Experts Connectors (MoECs) module to dynamically select experts tailored to different visual dimensions. To mitigate redundant or insufficient use of the visual information encoded by the MoECs module, we further design a Hierarchical Group Attention (HGA) with intra- and inter-group operations and an adaptive gating strategy for encoded visual features. We train MoCHA on two mainstream LLMs (e.g., Phi2-2.7B and Vicuna-7B) and evaluate their performance across various benchmarks. Notably, MoCHA outperforms state-of-the-art open-weight models on various tasks. For example, compared to CuMo (Mistral-7B), our MoCHA (Phi2-2.7B) presents outstanding abilities to mitigate hallucination by showing improvements of 3.25% in POPE and to follow visual instructions by raising 153 points on MME. Finally, ablation studies further confirm the effectiveness and robustness of the proposed MoECs and HGA in improving the overall performance of MoCHA.
Authors: Yuchen Zhou, Yan Luo, Xiaogang Wang, Xingjian Gu, Mingzhou Lu
Abstract: Efficient and high-accuracy 3D occupancy prediction is crucial for ensuring the performance of autonomous driving (AD) systems. However, many existing methods involve trade-offs between accuracy and efficiency. Some achieve high precision but with slow inference speed, while others adopt purely bird's-eye-view (BEV)-based 2D representations to accelerate processing, inevitably sacrificing vertical cues and compromising geometric integrity. To overcome these limitations, we propose a pure 2D framework that achieves efficient 3D occupancy prediction while preserving geometric integrity. Unlike conventional Lift-Splat-Shoot (LSS) methods that rely solely on depth scores to lift 2D features into 3D space, our approach additionally introduces a height-score projection to encode vertical geometric structure. We further employ direction-aware convolution to extract geometric features along both vertical and horizontal orientations, effectively balancing accuracy and computational efficiency. On the Occ3D-nuScenes, the proposed method achieves an mIoU of 39.3\% and an inference speed of 27.7 FPS, effectively balancing accuracy and efficiency. In simulations on edge devices, the inference speed reaches 14.8 FPS, further demonstrating the method's applicability for real-time deployment in resource-constrained environments.
Authors: Liang An, Jin Lyu, Li Lin, Pujin Cheng, Yebin Liu, Xiaoying Tang
Abstract: In the era of foundation models, achieving a unified understanding of different dynamic objects through a single network has the potential to empower stronger spatial intelligence. Moreover, accurate estimation of animal pose and shape across diverse species is essential for quantitative analysis in biological research. However, this topic remains underexplored due to the limited network capacity of previous methods and the scarcity of comprehensive multi-species datasets. To address these limitations, we introduce AniMer+, an extended version of our scalable AniMer framework. In this paper, we focus on a unified approach for reconstructing mammals (mammalia) and birds (aves). A key innovation of AniMer+ is its high-capacity, family-aware Vision Transformer (ViT) incorporating a Mixture-of-Experts (MoE) design. Its architecture partitions network layers into taxa-specific components (for mammalia and aves) and taxa-shared components, enabling efficient learning of both distinct and common anatomical features within a single model. To overcome the critical shortage of 3D training data, especially for birds, we introduce a diffusion-based conditional image generation pipeline. This pipeline produces two large-scale synthetic datasets: CtrlAni3D for quadrupeds and CtrlAVES3D for birds. To note, CtrlAVES3D is the first large-scale, 3D-annotated dataset for birds, which is crucial for resolving single-view depth ambiguities. Trained on an aggregated collection of 41.3k mammalian and 12.4k avian images (combining real and synthetic data), our method demonstrates superior performance over existing approaches across a wide range of benchmarks, including the challenging out-of-domain Animal Kingdom dataset. Ablation studies confirm the effectiveness of both our novel network architecture and the generated synthetic datasets in enhancing real-world application performance.
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: Qiaohui Chu, Haoyu Zhang, Meng Liu, Yisen Feng, Haoxiang Shi, Liqiang Nie
Abstract: Long-term action anticipation from egocentric video is critical for applications such as human-computer interaction and assistive technologies, where anticipating user intent enables proactive and context-aware AI assistance. However, existing approaches suffer from three key limitations: 1) underutilization of fine-grained visual cues from hand-object interactions, 2) neglect of semantic dependencies between verbs and nouns, and 3) lack of explicit cognitive reasoning, limiting generalization and long-term forecasting ability. To overcome these challenges, we propose INSIGHT, a unified two-stage framework for egocentric action anticipation. In the first stage, INSIGHT focuses on extracting semantically rich features from hand-object interaction regions and enhances action representations using a verb-noun co-occurrence matrix. In the second stage, it introduces a reinforcement learning-based module that simulates explicit cognitive reasoning through a structured process: visual perception (think) -> intention inference (reason) -> action anticipation (answer). Extensive experiments on Ego4D, EPIC-Kitchens-55, and EGTEA Gaze+ benchmarks show that INSIGHT achieves state-of-the-art performance, demonstrating its effectiveness and strong generalization capability.
Authors: Ziteng Wang, Siqi Yang, Limeng Qiao, Lin Ma
Abstract: Despite the success of Vision-Language Models (VLMs) like CLIP in aligning vision and language, their proficiency in detailed, fine-grained visual comprehension remains a key challenge. We present CLIP-IN, a novel framework that bolsters CLIP's fine-grained perception through two core innovations. Firstly, we leverage instruction-editing datasets, originally designed for image manipulation, as a unique source of hard negative image-text pairs. Coupled with a symmetric hard negative contrastive loss, this enables the model to effectively distinguish subtle visual-semantic differences. Secondly, CLIP-IN incorporates long descriptive captions, utilizing rotary positional encodings to capture rich semantic context often missed by standard CLIP. Our experiments demonstrate that CLIP-IN achieves substantial gains on the MMVP benchmark and various fine-grained visual recognition tasks, without compromising robust zero-shot performance on broader classification and retrieval tasks. Critically, integrating CLIP-IN's visual representations into Multimodal Large Language Models significantly reduces visual hallucinations and enhances reasoning abilities. This work underscores the considerable potential of synergizing targeted, instruction-based contrastive learning with comprehensive descriptive information to elevate the fine-grained understanding of VLMs.
Authors: Shuo Wang, Yongcai Wang, Zhaoxin Fan, Yucheng Wang, Maiyue Chen, Kaihui Wang, Zhizhong Su, Wanting Li, Xudong Cai, Yeying Jin, Deying Li
Abstract: Vision-Language Navigation (VLN) tasks often leverage panoramic RGB and depth inputs to provide rich spatial cues for action planning, but these sensors can be costly or less accessible in real-world deployments. Recent approaches based on Vision-Language Action (VLA) models achieve strong results with monocular input, yet they still lag behind methods using panoramic RGB-D information. We present MonoDream, a lightweight VLA framework that enables monocular agents to learn a Unified Navigation Representation (UNR). This shared feature representation jointly aligns navigation-relevant visual semantics (e.g., global layout, depth, and future cues) and language-grounded action intent, enabling more reliable action prediction. MonoDream further introduces Latent Panoramic Dreaming (LPD) tasks to supervise the UNR, which train the model to predict latent features of panoramic RGB and depth observations at both current and future steps based on only monocular input. Experiments on multiple VLN benchmarks show that MonoDream consistently improves monocular navigation performance and significantly narrows the gap with panoramic-based agents.
Authors: Sai Ma, Zhuang Li, John A Taylor
Abstract: Vision language models (VLMs) that enable natural language interaction with satellite imagery can democratize Earth observation by accelerating expert workflows, making data accessible to non-specialists, and enabling planet-scale automation. However, existing datasets focus mainly on short-term, high-resolution imagery from a limited number of satellites, overlooking low-resolution, multi-satellite, long-term archives, such as Landsat, that are essential for affordable and bias-robust global monitoring. We address this gap with Landsat30-AU, a large-scale vision-language dataset built from 30-meter resolution imagery collected by four Landsat satellites (5, 7, 8, and 9) over Australia, spanning more than 36 years. The dataset includes two components: Landsat30-AU-Cap, containing $196,262$ image-caption pairs, and Landsat30-AU-VQA, comprising 17,725 human-verified visual question answering (VQA) samples across eight remote sensing domains. Both datasets are curated through a bootstrapped pipeline that leverages generic VLMs with iterative refinement and human verification to ensure quality. Our evaluation of eight VLMs on our benchmark reveals that off-the-shelf models struggle to understand satellite imagery. The open-source remote-sensing VLM EarthDial achieves only 0.07 SPIDEr in captioning and a VQA accuracy of 0.48, highlighting the limitations of current approaches. Encouragingly, lightweight fine-tuning of Qwen2.5-VL-7B on Landsat30-AU improves captioning performance from 0.11 to 0.31 SPIDEr and boosts VQA accuracy from 0.74 to 0.87. Code and data are available at https://github.com/papersubmit1/landsat30-au.
Authors: Daisheng Jin, Ying He
Abstract: Reconstructing realistic 3D human avatars from monocular videos is a challenging task due to the limited geometric information and complex non-rigid motion involved. We present MonoCloth, a new method for reconstructing and animating clothed human avatars from monocular videos. To overcome the limitations of monocular input, we introduce a part-based decomposition strategy that separates the avatar into body, face, hands, and clothing. This design reflects the varying levels of reconstruction difficulty and deformation complexity across these components. Specifically, we focus on detailed geometry recovery for the face and hands. For clothing, we propose a dedicated cloth simulation module that captures garment deformation using temporal motion cues and geometric constraints. Experimental results demonstrate that MonoCloth improves both visual reconstruction quality and animation realism compared to existing methods. Furthermore, thanks to its part-based design, MonoCloth also supports additional tasks such as clothing transfer, underscoring its versatility and practical utility.
Authors: Young D. Kwon, Rui Li, Sijia Li, Da Li, Sourav Bhattacharya, Stylianos I. Venieris
Abstract: State-of-the-art text-to-image diffusion models (DMs) achieve remarkable quality, yet their massive parameter scale (8-11B) poses significant challenges for inferences on resource-constrained devices. In this paper, we present HierarchicalPrune, a novel compression framework grounded in a key observation: DM blocks exhibit distinct functional hierarchies, where early blocks establish semantic structures while later blocks handle texture refinements. HierarchicalPrune synergistically combines three techniques: (1) Hierarchical Position Pruning, which identifies and removes less essential later blocks based on position hierarchy; (2) Positional Weight Preservation, which systematically protects early model portions that are essential for semantic structural integrity; and (3) Sensitivity-Guided Distillation, which adjusts knowledge-transfer intensity based on our discovery of block-wise sensitivity variations. As a result, our framework brings billion-scale diffusion models into a range more suitable for on-device inference, while preserving the quality of the output images. Specifically, combined with INT4 weight quantisation, HierarchicalPrune achieves 77.5-80.4% memory footprint reduction (e.g., from 15.8 GB to 3.2 GB) and 27.9-38.0% latency reduction, measured on server and consumer grade GPUs, with the minimum drop of 2.6% in GenEval score and 7% in HPSv2 score compared to the original model. Finally, our comprehensive user study with 85 participants demonstrates that HierarchicalPrune maintains perceptual quality comparable to the original model while significantly outperforming prior works.
Authors: Xuan Wang, Kai Ruan, Liyang Qian, Zhizhi Guo, Chang Su, Gaoang Wang
Abstract: Text-driven motion generation has attracted increasing attention due to its broad applications in virtual reality, animation, and robotics. While existing methods typically model human and animal motion separately, a joint cross-species approach offers key advantages, such as a unified representation and improved generalization. However, morphological differences across species remain a key challenge, often compromising motion plausibility. To address this, we propose X-MoGen, the first unified framework for cross-species text-driven motion generation covering both humans and animals. X-MoGen adopts a two-stage architecture. First, a conditional graph variational autoencoder learns canonical T-pose priors, while an autoencoder encodes motion into a shared latent space regularized by morphological loss. In the second stage, we perform masked motion modeling to generate motion embeddings conditioned on textual descriptions. During training, a morphological consistency module is employed to promote skeletal plausibility across species. To support unified modeling, we construct UniMo4D, a large-scale dataset of 115 species and 119k motion sequences, which integrates human and animal motions under a shared skeletal topology for joint training. Extensive experiments on UniMo4D demonstrate that X-MoGen outperforms state-of-the-art methods on both seen and unseen species.
Authors: Can Zhao, Pengfei Guo, Dong Yang, Yucheng Tang, Yufan He, Benjamin Simon, Mason Belue, Stephanie Harmon, Baris Turkbey, Daguang Xu
Abstract: Medical image synthesis is an important topic for both clinical and research applications. Recently, diffusion models have become a leading approach in this area. Despite their strengths, many existing methods struggle with (1) limited generalizability that only work for specific body regions or voxel spacings, (2) slow inference, which is a common issue for diffusion models, and (3) weak alignment with input conditions, which is a critical issue for medical imaging. MAISI, a previously proposed framework, addresses generalizability issues but still suffers from slow inference and limited condition consistency. In this work, we present MAISI-v2, the first accelerated 3D medical image synthesis framework that integrates rectified flow to enable fast and high quality generation. To further enhance condition fidelity, we introduce a novel region-specific contrastive loss to enhance the sensitivity to region of interest. Our experiments show that MAISI-v2 can achieve SOTA image quality with $33 \times$ acceleration for latent diffusion model. We also conducted a downstream segmentation experiment to show that the synthetic images can be used for data augmentation. We release our code, training details, model weights, and a GUI demo to facilitate reproducibility and promote further development within the community.
Authors: Weichen Zhang, Kebin Liu, Fan Dang, Zhui Zhu, Xikai Sun, Yunhao Liu
Abstract: Semantic segmentation in open-vocabulary scenarios presents significant challenges due to the wide range and granularity of semantic categories. Existing weakly-supervised methods often rely on category-specific supervision and ill-suited feature construction methods for contrastive learning, leading to semantic misalignment and poor performance. In this work, we propose a novel weakly-supervised approach, SynSeg, to address the challenges. SynSeg performs Multi-Category Contrastive Learning (MCCL) as a stronger training signal with a new feature reconstruction framework named Feature Synergy Structure (FSS). Specifically, MCCL strategy robustly combines both intra- and inter-category alignment and separation in order to make the model learn the knowledge of correlations from different categories within the same image. Moreover, FSS reconstructs discriminative features for contrastive learning through prior fusion and semantic-activation-map enhancement, effectively avoiding the foreground bias introduced by the visual encoder. Furthermore, SynSeg is a lightweight end-to-end solution without using any mid-term output from large-scale pretrained models and capable for real-time inference. In general, SynSeg effectively improves the abilities in semantic localization and discrimination under weak supervision in an efficient manner. Extensive experiments on benchmarks demonstrate that our method outperforms state-of-the-art (SOTA) performance. Particularly, SynSeg achieves higher accuracy than SOTA baselines with a ratio from 6.9\% up to 26.2\%.
Authors: Andrii Yermakov, Jan Cech, Jiri Matas, Mario Fritz
Abstract: The generalization of deepfake detectors to unseen manipulation techniques remains a challenge for practical deployment. Although many approaches adapt foundation models by introducing significant architectural complexity, this work demonstrates that robust generalization is achievable through a parameter-efficient adaptation of one of the foundational pre-trained vision encoders. The proposed method, GenD, fine-tunes only the Layer Normalization parameters (0.03% of the total) and enhances generalization by enforcing a hyperspherical feature manifold using L2 normalization and metric learning on it. We conducted an extensive evaluation on 14 benchmark datasets spanning from 2019 to 2025. The proposed method achieves state-of-the-art performance, outperforming more complex, recent approaches in average cross-dataset AUROC. Our analysis yields two primary findings for the field: 1) training on paired real-fake data from the same source video is essential for mitigating shortcut learning and improving generalization, and 2) detection difficulty on academic datasets has not strictly increased over time, with models trained on older, diverse datasets showing strong generalization capabilities. This work delivers a computationally efficient and reproducible method, proving that state-of-the-art generalization is attainable by making targeted, minimal changes to a pre-trained foundational image encoder model. The code is at: https://github.com/yermandy/GenD
Authors: Junsheng Huang, Shengyu Hao, Bocheng Hu, Hongwei Wang, Gaoang Wang
Abstract: Understanding dynamic 4D scenes from an egocentric perspective-modeling changes in 3D spatial structure over time-is crucial for human-machine interaction, autonomous navigation, and embodied intelligence. While existing egocentric datasets contain dynamic scenes, they lack unified 4D annotations and task-driven evaluation protocols for fine-grained spatio-temporal reasoning, especially on motion of objects and human, together with their interactions. To address this gap, we introduce EgoDynamic4D, a novel QA benchmark on highly dynamic scenes, comprising RGB-D video, camera poses, globally unique instance masks, and 4D bounding boxes. We construct 927K QA pairs accompanied by explicit Chain-of-Thought (CoT), enabling verifiable, step-by-step spatio-temporal reasoning. We design 12 dynamic QA tasks covering agent motion, human-object interaction, trajectory prediction, relation understanding, and temporal-causal reasoning, with fine-grained, multidimensional metrics. To tackle these tasks, we propose an end-to-end spatio-temporal reasoning framework that unifies dynamic and static scene information, using instance-aware feature encoding, time and camera encoding, and spatially adaptive down-sampling to compress large 4D scenes into token sequences manageable by LLMs. Experiments on EgoDynamic4D show that our method consistently outperforms baselines, validating the effectiveness of multimodal temporal modeling for egocentric dynamic scene understanding.
Authors: Reuben Dorent, Laura Rigolo, Colin P. Galvin, Junyu Chen, Mattias P. Heinrich, Aaron Carass, Olivier Colliot, Demian Wassermann, Alexandra Golby, Tina Kapur, William Wells
Abstract: Accurate intraoperative image guidance is critical for achieving maximal safe resection in brain tumor surgery, yet neuronavigation systems based on preoperative MRI lose accuracy during the procedure due to brain shift. Aligning post-resection intraoperative ultrasound (iUS) with preoperative MRI can restore spatial accuracy by estimating brain shift deformations, but it remains a challenging problem given the large anatomical and topological changes and substantial modality intensity gap. The ReMIND2Reg 2025 Challenge provides the largest public benchmark for this task, built upon the ReMIND dataset. It offers 99 training cases, 5 validation cases, and 10 private test cases comprising paired 3D ceT1 MRI, T2 MRI, and post-resection 3D iUS volumes. Data are provided without annotations for training, while validation and test performance are evaluated on manually annotated anatomical landmarks. Metrics include target registration error (TRE), robustness to worst-case landmark misalignment (TRE30), and runtime. By establishing a standardized evaluation framework for this clinically critical and technically complex problem, ReMIND2Reg aims to accelerate the development of robust, generalizable, and clinically deployable multimodal registration algorithms for image-guided neurosurgery.
Authors: Rajan Das Gupta, Md Yeasin Rahat, Nafiz Fahad, Abir Ahmed, Liew Tze Hui
Abstract: This study investigates how large language models (LLMs) can be used to understand human behavior using motion and video data. We think that mixing both types is essential to completely capture the nuanced movements and meanings of human actions, in contrast to recent models that simply concentrate on motion data or films. To address this, we provide ViMoNet, a straightforward yet effective framework for comprehending, characterizing, and deducing human action. ViMoNet employs a joint training strategy that leverages the advantages of two data types: detailed motion-text data, which is more exact, and generic video-text data, which is more comprehensive but less detailed. This aids in the model's acquisition of rich data regarding time and space in human behavior. Additionally, we provide a brand new dataset named VIMOS that contains a variety of films, motion sequences, instructions, and subtitles. We developed ViMoNet-Bench, a standardized benchmark with carefully labeled samples, to evaluate how well models understand human behavior. Our tests show that ViMoNet outperforms existing methods in caption generation, motion understanding, and behavior interpretation.
Authors: Chengtao Lv, Bilang Zhang, Yang Yong, Ruihao Gong, Yushi Huang, Shiqiao Gu, Jiajun Wu, Yumeng Shi, Jinyang Guo, Wenya Wang
Abstract: Large Vision-Language Models (VLMs) exhibit impressive multi-modal capabilities but suffer from prohibitive computational and memory demands, due to their long visual token sequences and massive parameter sizes. To address these issues, recent works have proposed training-free compression methods. However, existing efforts often suffer from three major limitations: (1) Current approaches do not decompose techniques into comparable modules, hindering fair evaluation across spatial and temporal redundancy. (2) Evaluation confined to simple single-turn tasks, failing to reflect performance in realistic scenarios. (3) Isolated use of individual compression techniques, without exploring their joint potential. To overcome these gaps, we introduce LLMC+, a comprehensive VLM compression benchmark with a versatile, plug-and-play toolkit. LLMC+ supports over 20 algorithms across five representative VLM families and enables systematic study of token-level and model-level compression. Our benchmark reveals that: (1) Spatial and temporal redundancies demand distinct technical strategies. (2) Token reduction methods degrade significantly in multi-turn dialogue and detail-sensitive tasks. (3) Combining token and model compression achieves extreme compression with minimal performance loss. We believe LLMC+ will facilitate fair evaluation and inspire future research in efficient VLM. Our code is available at https://github.com/ModelTC/LightCompress.
Authors: Yangjie Xiao, Ke Zhang, Jiacun Wang, Xin Sheng, Yurong Guo, Meijuan Chen, Zehua Ren, Zhaoye Zheng, Zhenbing Zhao
Abstract: Bolt defect detection is critical to ensure the safety of transmission lines. However, the scarcity of defect images and imbalanced data distributions significantly limit detection performance. To address this problem, we propose a segmentationdriven bolt defect editing method (SBDE) to augment the dataset. First, a bolt attribute segmentation model (Bolt-SAM) is proposed, which enhances the segmentation of complex bolt attributes through the CLAHE-FFT Adapter (CFA) and Multipart- Aware Mask Decoder (MAMD), generating high-quality masks for subsequent editing tasks. Second, a mask optimization module (MOD) is designed and integrated with the image inpainting model (LaMa) to construct the bolt defect attribute editing model (MOD-LaMa), which converts normal bolts into defective ones through attribute editing. Finally, an editing recovery augmentation (ERA) strategy is proposed to recover and put the edited defect bolts back into the original inspection scenes and expand the defect detection dataset. We constructed multiple bolt datasets and conducted extensive experiments. Experimental results demonstrate that the bolt defect images generated by SBDE significantly outperform state-of-the-art image editing models, and effectively improve the performance of bolt defect detection, which fully verifies the effectiveness and application potential of the proposed method. The code of the project is available at https://github.com/Jay-xyj/SBDE.
Authors: Zheng Qin, Ruobing Zheng, Yabing Wang, Tianqi Li, Yi Yuan, Jingdong Chen, Le Wang
Abstract: While Multimodal Large Language Models (MLLMs) show immense promise for achieving truly human-like interactions, progress is hindered by the lack of fine-grained evaluation frameworks for human-centered scenarios, encompassing both the understanding of complex human intentions and the provision of empathetic, context-aware responses. Here we introduce HumanSense, a comprehensive benchmark designed to evaluate the human-centered perception and interaction capabilities of MLLMs, with a particular focus on deep understanding of extended multimodal contexts and the formulation of rational feedback. Our evaluation reveals that leading MLLMs still have considerable room for improvement, particularly for advanced interaction-oriented tasks. Supplementing visual input with audio and text information yields substantial improvements, and Omni-modal models show advantages on these tasks.Furthermore, grounded in the observation that appropriate feedback stems from a contextual analysis of the interlocutor's needs and emotions, we posit that reasoning ability serves as the key to unlocking it. We devise a multi-stage, modality-progressive reinforcement learning approach, resulting in HumanSense-Omni-Reasoning, which substantially enhances performance on higher-level understanding and interactive tasks. Additionally, we observe that successful reasoning processes appear to exhibit consistent thought patterns. By designing corresponding prompts, we also enhance the performance of non-reasoning models in a training-free manner.Project page: \textcolor{brightpink}{https://digital-avatar.github.io/ai/HumanSense/}
Authors: Md Asgor Hossain Reaj, Rajan Das Gupta, Md Yeasin Rahat, Nafiz Fahad, Md Jawadul Hasan, Tze Hui Liew
Abstract: We introduce GANDiff FR, the first synthetic framework that precisely controls demographic and environmental factors to measure, explain, and reduce bias with reproducible rigor. GANDiff FR unifies StyleGAN3-based identity-preserving generation with diffusion-based attribute control, enabling fine-grained manipulation of pose around 30 degrees, illumination (four directions), and expression (five levels) under ceteris paribus conditions. We synthesize 10,000 demographically balanced faces across five cohorts validated for realism via automated detection (98.2%) and human review (89%) to isolate and quantify bias drivers. Benchmarking ArcFace, CosFace, and AdaFace under matched operating points shows AdaFace reduces inter-group TPR disparity by 60% (2.5% vs. 6.3%), with illumination accounting for 42% of residual bias. Cross-dataset evaluation on RFW, BUPT, and CASIA WebFace confirms strong synthetic-to-real transfer (r 0.85). Despite around 20% computational overhead relative to pure GANs, GANDiff FR yields three times more attribute-conditioned variants, establishing a reproducible, regulation-aligned (EU AI Act) standard for fairness auditing. Code and data are released to support transparent, scalable bias evaluation.
Authors: Yumeng Shi, Quanyu Long, Yin Wu, Wenya Wang
Abstract: Video language models (VideoLMs) have made significant progress in multimodal understanding. However, temporal understanding, which involves identifying event order, duration, and relationships across time, still remains a core challenge. Prior works emphasize positional encodings (PEs) as a key mechanism for encoding temporal structure. Surprisingly, we find that removing or modifying PEs in video inputs yields minimal degradation in the performance of temporal understanding. In contrast, reversing the frame sequence while preserving the original PEs causes a substantial drop. To explain this behavior, we conduct substantial analysis experiments to trace how temporal information is integrated within the model. We uncover a causal information pathway: temporal cues are progressively synthesized through inter-frame attention, aggregated in the final frame, and subsequently integrated into the query tokens. This emergent mechanism shows that temporal reasoning emerges from inter-visual token interactions under the constraints of causal attention, which implicitly encodes temporal structure. Based on these insights, we propose two efficiency-oriented strategies: staged cross-modal attention and a temporal exit mechanism for early token truncation. Experiments on two benchmarks validate the effectiveness of both approaches. To the best of our knowledge, this is the first systematic study of video temporal understanding in VideoLMs, offering insights for future model improvement. Our code is available at https://github.com/ANDgate99/Causality-Matters .
Authors: Liang Lv, Di Wang, Jing Zhang, Lefei Zhang
Abstract: Semi-supervised semantic segmentation (S4) has advanced remote sensing (RS) analysis by leveraging unlabeled data through pseudo-labeling and consistency learning. However, existing S4 studies often rely on small-scale datasets and models, limiting their practical applicability. To address this, we propose S5, the first scalable framework for semi-supervised semantic segmentation in RS, which unlocks the potential of vast unlabeled Earth observation data typically underutilized due to costly pixel-level annotations. Built upon existing large-scale RS datasets, S5 introduces a data selection strategy that integrates entropy-based filtering and diversity expansion, resulting in the RS4P-1M dataset. Using this dataset, we systematically scale up S4 methods by pre-training RS foundation models (RSFMs) of varying sizes on this extensive corpus, significantly boosting their performance on land cover segmentation and object detection tasks. Furthermore, during fine-tuning, we incorporate a Mixture-of-Experts (MoE)-based multi-dataset fine-tuning approach, which enables efficient adaptation to multiple RS benchmarks with fewer parameters. This approach improves the generalization and versatility of RSFMs across diverse RS benchmarks. The resulting RSFMs achieve state-of-the-art performance across all benchmarks, underscoring the viability of scaling semi-supervised learning for RS applications. All datasets, code, and models will be released at https://github.com/MiliLab/S5
Authors: Chen Qian, Xinran Yu, Zewen Huang, Danyang Li, Qiang Ma, Fan Dang, Xuan Ding, Guangyong Shang, Zheng Yang
Abstract: Vision-Language Models (VLMs) are increasingly deployed in real-time applications such as autonomous driving and human-computer interaction, which demand fast and reliable responses based on accurate perception. To meet these requirements, existing systems commonly employ cloud-edge collaborative architectures, such as partitioned Large Vision-Language Models (LVLMs) or task offloading strategies between Large and Small Vision-Language Models (SVLMs). However, these methods fail to accommodate cloud latency fluctuations and overlook the full potential of delayed but accurate LVLM responses. In this work, we propose a novel cloud-edge collaborative paradigm for VLMs, termed Context Transfer, which treats the delayed outputs of LVLMs as historical context to provide real-time guidance for SVLMs inference. Based on this paradigm, we design edgeVLM, which incorporates both context replacement and visual focus modules to refine historical textual input and enhance visual grounding consistency. Extensive experiments on three real-time vision-lanuage reasoning tasks across four datasets demonstrate the effectiveness of the proposed framework. The new paradigm lays the groundwork for more effective and latency-aware collaboration strategies in future VLM systems.
Authors: Maryam Adelipour, Gustavo Carneiro, Jeongkwon Kim
Abstract: Sebocytes are lipid-secreting cells whose differentiation is marked by the accumulation of intracellular lipid droplets, making their quantification a key readout in sebocyte biology. Manual counting is labor-intensive and subjective, motivating automated solutions. Here, we introduce a simple attention-based multiple instance learning (MIL) framework for sebocyte image analysis. Nile Red-stained sebocyte images were annotated into 14 classes according to droplet counts, expanded via data augmentation to about 50,000 cells. Two models were benchmarked: a baseline multi-layer perceptron (MLP) trained on aggregated patch-level counts, and an attention-based MIL model leveraging ResNet-50 features with instance weighting. Experiments using five-fold cross-validation showed that the baseline MLP achieved more stable performance (mean MAE = 5.6) compared with the attention-based MIL, which was less consistent (mean MAE = 10.7) but occasionally superior in specific folds. These findings indicate that simple bag-level aggregation provides a robust baseline for slide-level droplet counting, while attention-based MIL requires task-aligned pooling and regularization to fully realize its potential in sebocyte image analysis.
Authors: Himanshu Singh, A. V. Subramanyam, Shivank Rajput, Mohan Kankanhalli
Abstract: Deep neural networks have exhibited impressive performance in image classification tasks but remain vulnerable to adversarial examples. Standard adversarial training enhances robustness but typically fails to explicitly address inter-class feature overlap, a significant contributor to adversarial susceptibility. In this work, we introduce a novel adversarial training framework that actively mitigates inter-class proximity by projecting out inter-class dependencies from adversarial and clean samples in the feature space. Specifically, our approach first identifies the nearest inter-class neighbors for each adversarial sample and subsequently removes projections onto these neighbors to enforce stronger feature separability. Theoretically, we demonstrate that our proposed logits correction reduces the Lipschitz constant of neural networks, thereby lowering the Rademacher complexity, which directly contributes to improved generalization and robustness. Extensive experiments across standard benchmarks including CIFAR-10, CIFAR-100, and SVHN show that our method demonstrates strong performance that is competitive with leading adversarial training techniques, highlighting significant achievements in both robust and clean accuracy. Our findings reveal the importance of addressing inter-class feature proximity explicitly to bolster adversarial robustness in DNNs.
Authors: Tiancheng Yang, Lin Zhang, Jiaye Lin, Guimin Hu, Di Wang, Lijie Hu
Abstract: Multimodal Large Language Models (MLLMs) achieve strong performance on tasks like image captioning and visual question answering, but remain prone to hallucinations, where generated text conflicts with the visual input. Prior work links this partly to insufficient visual attention, but existing attention-based detectors and mitigation typically apply uniform adjustments across layers and heads, obscuring where errors originate. In this paper, we first show these methods fail to accurately localize problematic layers. Then, we introduce two diagnostics: Layer Image Attention Entropy (LIAE) which flags anomalous layers, and Image Attention Focus (IAF) which scores attention heads within those layers. Analysis shows that LIAE pinpoints faulty layers and IAF reliably ranks heads that warrant correction. Guided by these signals, we propose Dynamic Layer-wise Entropy and Attention Fusion (D-LEAF), a task-agnostic, attention-guided method that dynamically localizes and corrects errors during inference with negligible overhead. Furthermore, by establishing a connection between D-LEAF and DPO, we provide theoretical justification for the effectiveness of D-LEAF. Results show our D-LEAF delivers a 53\% relative improvement on standard captioning benchmarks, and on VQA both accuracy and F1-score improve by approximately 4\%, substantially suppressing hallucinations while preserving efficiency.
Authors: Zhongyu Xia, Hansong Yang, Yongtao Wang
Abstract: Three-dimensional Object Detection from multi-view cameras and LiDAR is a crucial component for autonomous driving and smart transportation. However, in the process of basic feature extraction, perspective transformation, and feature fusion, noise and error will gradually accumulate. To address this issue, we propose InsFusion, which can extract proposals from both raw and fused features and utilizes these proposals to query the raw features, thereby mitigating the impact of accumulated errors. Additionally, by incorporating attention mechanisms applied to the raw features, it thereby mitigates the impact of accumulated errors. Experiments on the nuScenes dataset demonstrate that InsFusion is compatible with various advanced baseline methods and delivers new state-of-the-art performance for 3D object detection.
Authors: Zongliang Wu, Siming Zheng, Peng-Tao Jiang, Xin Yuan
Abstract: Pre-trained diffusion models have shown great potential in real-world image super-resolution (Real-ISR) tasks by enabling high-resolution reconstructions. While one-step diffusion (OSD) methods significantly improve efficiency compared to traditional multi-step approaches, they still have limitations in balancing fidelity and realism across diverse scenarios. Since the OSDs for SR are usually trained or distilled by a single timestep, they lack flexible control mechanisms to adaptively prioritize these competing objectives, which are inherently manageable in multi-step methods through adjusting sampling steps. To address this challenge, we propose a Realism Controlled One-step Diffusion (RCOD) framework for Real-ISR. RCOD provides a latent domain grouping strategy that enables explicit control over fidelity-realism trade-offs during the noise prediction phase with minimal training paradigm modifications and original training data. A degradation-aware sampling strategy is also introduced to align distillation regularization with the grouping strategy and enhance the controlling of trade-offs. Moreover, a visual prompt injection module is used to replace conventional text prompts with degradation-aware visual tokens, enhancing both restoration accuracy and semantic consistency. Our method achieves superior fidelity and perceptual quality while maintaining computational efficiency. Extensive experiments demonstrate that RCOD outperforms state-of-the-art OSD methods in both quantitative metrics and visual qualities, with flexible realism control capabilities in the inference stage.
Authors: Navid Hashemi, Samuel Sasaki, Diego Manzanas Lopez, Lars Lindemann, Ipek Oguz, Meiyi Ma, Taylor T. Johnson
Abstract: Semantic segmentation networks (SSNs) are central to safety-critical applications such as medical imaging and autonomous driving, where robustness under uncertainty is essential. However, existing probabilistic verification methods often fail to scale with the complexity and dimensionality of modern segmentation tasks, producing guarantees that are overly conservative and of limited practical value. We propose a probabilistic verification framework that is architecture-agnostic and scalable to high-dimensional input-output spaces. Our approach employs conformal inference (CI), enhanced by a novel technique that we call the \textbf{clipping block}, to provide provable guarantees while mitigating the excessive conservatism of prior methods. Experiments on large-scale segmentation models across CamVid, OCTA-500, Lung Segmentation, and Cityscapes demonstrate that our framework delivers reliable safety guarantees while substantially reducing conservatism compared to state-of-the-art approaches on segmentation tasks. We also provide a public GitHub repository (https://github.com/Navidhashemicodes/SSN_Reach_CLP_Surrogate) for this approach, to support reproducibility.
URLs: https://github.com/Navidhashemicodes/SSN_Reach_CLP_Surrogate)
Authors: Bingyu Li, Haocheng Dong, Da Zhang, Zhiyuan Zhao, Junyu Gao, Xuelong Li
Abstract: Open-Vocabulary Remote Sensing Image Segmentation (OVRSIS), an emerging task that adapts Open-Vocabulary Segmentation (OVS) to the remote sensing (RS) domain, remains underexplored due to the absence of a unified evaluation benchmark and the domain gap between natural and RS images. To bridge these gaps, we first establish a standardized OVRSIS benchmark (\textbf{OVRSISBench}) based on widely-used RS segmentation datasets, enabling consistent evaluation across methods. Using this benchmark, we comprehensively evaluate several representative OVS/OVRSIS models and reveal their limitations when directly applied to remote sensing scenarios. Building on these insights, we propose \textbf{RSKT-Seg}, a novel open-vocabulary segmentation framework tailored for remote sensing. RSKT-Seg integrates three key components: (1) a Multi-Directional Cost Map Aggregation (RS-CMA) module that captures rotation-invariant visual cues by computing vision-language cosine similarities across multiple directions; (2) an Efficient Cost Map Fusion (RS-Fusion) transformer, which jointly models spatial and semantic dependencies with a lightweight dimensionality reduction strategy; and (3) a Remote Sensing Knowledge Transfer (RS-Transfer) module that injects pre-trained knowledge and facilitates domain adaptation via enhanced upsampling. Extensive experiments on the benchmark show that RSKT-Seg consistently outperforms strong OVS baselines by +3.8 mIoU and +5.9 mACC, while achieving 2x faster inference through efficient aggregation. Our code is \href{https://github.com/LiBingyu01/RSKT-Seg}{\textcolor{blue}{here}}.
Authors: Hao Wang, Eiki Murata, Lingfang Zhang, Ayako Sato, So Fukuda, Ziqi Yin, Wentao Hu, Keisuke Nakao, Yusuke Nakamura, Sebastian Zwirner, Yi-Chia Chen, Hiroyuki Otomo, Hiroki Ouchi, Daisuke Kawahara
Abstract: Recent advances in multimodal large language models (MLLMs) have significantly enhanced video understanding capabilities, opening new possibilities for practical applications. Yet current video benchmarks focus largely on indoor scenes or short-range outdoor activities, leaving the challenges associated with long-distance travel largely unexplored. Mastering extended geospatial-temporal trajectories is critical for next-generation MLLMs, underpinning real-world tasks such as embodied-AI planning and navigation. To bridge this gap, we present VIR-Bench, a novel benchmark consisting of 200 travel videos that frames itinerary reconstruction as a challenging task designed to evaluate and push forward MLLMs' geospatial-temporal intelligence. Experimental results reveal that state-of-the-art MLLMs, including proprietary ones, struggle to achieve high scores, underscoring the difficulty of handling videos that span extended spatial and temporal scales. Moreover, we conduct an in-depth case study in which we develop a prototype travel-planning agent that leverages the insights gained from VIR-Bench. The agent's markedly improved itinerary recommendations verify that our evaluation protocol not only benchmarks models effectively but also translates into concrete performance gains in user-facing applications.
Authors: Xu Yang, Gady Agam
Abstract: Multi-object tracking (MOT) in videos remains challenging due to complex object motions and crowded scenes. Recent DETR-based frameworks offer end-to-end solutions but typically process detection and tracking queries jointly within a single Transformer Decoder layer, leading to conflicts and degraded association accuracy. We introduce the Motion-Aware Transformer (MATR), which explicitly predicts object movements across frames to update track queries in advance. By reducing query collisions, MATR enables more consistent training and improves both detection and association. Extensive experiments on DanceTrack, SportsMOT, and BDD100k show that MATR delivers significant gains across standard metrics. On DanceTrack, MATR improves HOTA by more than 9 points over MOTR without additional data and reaches a new state-of-the-art score of 71.3 with supplementary data. MATR also achieves state-of-the-art results on SportsMOT (72.2 HOTA) and BDD100k (54.7 mTETA, 41.6 mHOTA) without relying on external datasets. These results demonstrate that explicitly modeling motion within end-to-end Transformers offers a simple yet highly effective approach to advancing multi-object tracking.
Authors: Mohsen Ghafoorian, Denis Korzhenkov, Amirhossein Habibian
Abstract: Transformer-based video diffusion models (VDMs) deliver state-of-the-art video generation quality but are constrained by the quadratic cost of self-attention, making long sequences and high resolutions computationally expensive. While linear attention offers sub-quadratic complexity, previous approaches have failed to match the expressiveness of softmax attention unless retrained at significant computational cost. We introduce Attention Surgery, an efficient framework that enables linear or hybrid attention in pretrained VDMs, eliminating the need for training from scratch. Inspired by recent advances in language models, our method combines a novel hybrid attention mechanism-mixing softmax and linear tokens-with a lightweight distillation and fine-tuning pipeline requiring only a few GPU-days. Additionally, we incorporate a cost-aware block-rate strategy to balance expressiveness and efficiency across layers. Applied to Wan2.1 1.3B, a state-of-the-art efficient transformer VDM and evaluated on VBench, VBench2.0 and a human preference study, Attention Surgery achieves competitive results. Furthermore, measurements of on-mobile latency, memory usage, and FLOPs demonstrate notable improvements in scaling behavior for longer videos. Project page is available at: https://qualcomm-ai-research.github.io/attention-surgery.
URLs: https://qualcomm-ai-research.github.io/attention-surgery.
Authors: Ran Tong, Jiaqi Liu, Tong Wang, Xin Hu, Su Liu, Lanruo Wang, Jiexi Xu
Abstract: The accurate interpretation of chest radiographs using automated methods is a critical task in medical imaging. This paper presents a comparative analysis between a supervised lightweight Convolutional Neural Network (CNN) and a state-of-the-art, zero-shot medical Vision-Language Model (VLM), BiomedCLIP, across two distinct diagnostic tasks: pneumonia detection on the PneumoniaMNIST benchmark and tuberculosis detection on the Shenzhen TB dataset. Our experiments show that supervised CNNs serve as highly competitive baselines in both cases. While the default zero-shot performance of the VLM is lower, we demonstrate that its potential can be unlocked via a simple yet crucial remedy: decision threshold calibration. By optimizing the classification threshold on a validation set, the performance of BiomedCLIP is significantly boosted across both datasets. For pneumonia detection, calibration enables the zero-shot VLM to achieve a superior F1-score of 0.8841, surpassing the supervised CNN's 0.8803. For tuberculosis detection, calibration dramatically improves the F1-score from 0.4812 to 0.7684, bringing it close to the supervised baseline's 0.7834. This work highlights a key insight: proper calibration is essential for leveraging the full diagnostic power of zero-shot VLMs, enabling them to match or even outperform efficient, task-specific supervised models.
Authors: Matthias Perkonigg, Patrick Rockenschaub, Georg G\"obel, Adelheid W\"ohrer
Abstract: Accurate brain tumor classification is critical for intra-operative decision making in neuro-oncological surgery. However, existing approaches are restricted to a fixed set of predefined classes and are therefore unable to capture patterns of tumor types not available during training. Unsupervised learning can extract general-purpose features, but it lacks the ability to incorporate prior knowledge from labelled data, and semi-supervised methods often assume that all potential classes are represented in the labelled data. Generalized Category Discovery (GCD) aims to bridge this gap by categorizing both known and unknown classes within unlabelled data. To reflect the hierarchical structure of brain tumor taxonomies, in this work, we introduce Hierarchical Generalized Category Discovery for Brain Tumor Classification (HGCD-BT), a novel approach that integrates hierarchical clustering with contrastive learning. Our method extends contrastive learning based GCD by incorporating a novel semi-supervised hierarchical clustering loss. We evaluate HGCD-BT on OpenSRH, a dataset of stimulated Raman histology brain tumor images, achieving a +28% improvement in accuracy over state-of-the-art GCD methods for patch-level classification, particularly in identifying previously unseen tumor categories. Furthermore, we demonstrate the generalizability of HGCD-BT on slide-level classification of hematoxylin and eosin stained whole-slide images from the Digital Brain Tumor Atlas, confirming its utility across imaging modalities.
Authors: Python Song, Luke Tenyi Chang, Yun-Yun Tsai, Penghui Li, Junfeng Yang
Abstract: CAPTCHA, originally designed to distinguish humans from robots, has evolved into a real-world benchmark for assessing the spatial reasoning capabilities of vision-language models. In this work, we first show that step-by-step reasoning is crucial for vision-language models (VLMs) to solve CAPTCHAs, which represent high-difficulty spatial reasoning tasks, and that current commercial vision-language models still struggle with such reasoning. In particular, we observe that most commercial VLMs (e.g., Gemini, Claude, GPT, etc.) fail to effectively solve CAPTCHAs and thus achieve low accuracy (around 21.9 percent). However, our findings indicate that requiring the model to perform step-by-step reasoning before generating the final coordinates can significantly enhance its solving accuracy, underscoring the severity of the gap. To systematically study this issue, we introduce CAPTCHA-X, the first real-world CAPTCHA benchmark with reasoning, covering seven categories of CAPTCHAs (such as Gobang, hCaptcha, etc.) with step-by-step action solutions and grounding annotations. We further define five reasoning-oriented metrics that enable a comprehensive evaluation of models reasoning capabilities. To validate the effectiveness of reasoning, we also propose a general agentic VLM-based framework that incorporates the models inherent reasoning abilities. Our method achieves state-of-the-art performance across five high-difficulty CAPTCHA types, with an average solving accuracy of 83.9 percent, substantially surpassing existing baselines. These results reveal the limitations of current models and highlight the importance of reasoning in advancing visual-spatial challenges in the future.
Authors: Ieva Bagdonaviciute, Vibhav Vineet
Abstract: While recent Vision-Language Models (VLMs) have achieved impressive progress, it remains difficult to determine why they succeed or fail on complex reasoning tasks. Traditional benchmarks evaluate what models can answer correctly, not why they succeed or fail. In this work, we perform a failure-mode analysis of six frontier VLMs on three physics-based benchmarks - Physion, Physion++, and CLEVRER - by introducing custom subtests (for Physion and Physion++) and an integration of existing benchmark categories (for CLEVRER) to factor benchmark performance into distinct, testable capabilities. These subtests isolate perception (object, color, and occlusion recognition) and physics understanding (motion prediction and spatial reasoning), enabling us to test whether models attend to the correct entities and dynamics underlying their answers. Counterintuitively, subtest mastery correlates only weakly with benchmark accuracy: models often answer correctly without grounding in perception or physics. This suggests that current VLMs sometimes achieve benchmark scores for the wrong reasons, underscoring the need for diagnostics that expose hidden failure modes beyond aggregate metrics.
Authors: Zhihao Wen, Wenkang Wei, Yuan Fang, Xingtong Yu, Hui Zhang, Weicheng Zhu, Xin Zhang
Abstract: Knowledge-based Visual Question Answering (KVQA) requires models to ground entities in images and reason over factual knowledge. Recent work has introduced its implicit-knowledge variant, IK-KVQA, where a multimodal large language model (MLLM) is the sole knowledge source and answers are produced without external retrieval. Existing IK-KVQA approaches, however, are typically trained with answer-only supervision: reasoning remains implicit, justifications are often weak or inconsistent, and generalization after standard supervised fine-tuning (SFT) can be brittle. We propose MODELNAME, a framework that equips IK-KVQA with dual-path structured reasoning traces (symbolic relation paths over text and vision together with path-grounded natural-language explanations) to provide a stronger inductive bias than generic answer-only supervision. These traces act as modality-aware scaffolds that guide the model toward relevant entities and attributes, offering more structure than generic chain-of-thought supervision while not constraining reasoning to any single fixed path. Using a single open-source MLLM, MODELNAME constructs and selects traces to build an offline trace-enriched dataset and then performs structure-aware self-distillation; no external retrievers, verifiers, or curated knowledge bases are used, and inference is a single autoregressive pass. Across benchmarks, MODELNAME consistently improves both answer accuracy and the transparency of intermediate reasoning, achieving up to 11.3% higher answer accuracy on OK-VQA over the strongest baseline.
Authors: James Baker
Abstract: Personalizing diffusion models allows users to generate new images that incorporate a given subject, allowing more control than a text prompt. These models often suffer somewhat when they end up just recreating the subject image and ignoring the text prompt. We observe that one popular method for personalization, IP-Adapter, automatically generates masks that segment the subject from the background during inference. We propose to use this automatically generated mask on a second pass to mask the image tokens, thus restricting them to the subject, not the background, allowing the text prompt to attend to the rest of the image. For text prompts describing locations and places, this produces images that accurately depict the subject while definitively matching the prompt. We compare our method to a few other test time personalization methods, and find our method displays high prompt and source image alignment. We also perform a user study to validate whether end users would appreciate our method. Code available at https://github.com/jamesBaker361/monkey
Authors: Shuhei Tarashima, Yushan Wang, Norio Tagawa
Abstract: In this work, we aim to develop simple and efficient models for human mesh recovery (HMR) and its predecessor task, human pose estimation (HPE). State-of-the-art HMR methods, such as HMR2.0 and its successors, rely on large, non-hierarchical vision transformers as encoders, which are inherited from the corresponding HPE models like ViTPose. To establish baselines across varying computational budgets, we first construct three lightweight HMR2.0 variants by adapting the corresponding ViTPose models. In addition, we propose leveraging the early stages of hierarchical vision foundation models (VFMs), including Swin Transformer, GroupMixFormer, and VMamba, as encoders. This design is motivated by the observation that intermediate stages of hierarchical VFMs produce feature maps with resolutions comparable to or higher than those of non-hierarchical counterparts. We conduct a comprehensive evaluation of 27 hierarchical-VFM-based HMR and HPE models, demonstrating that using only the first two or three stages achieves performance on par with full-stage models. Moreover, we show that the resulting truncated models exhibit better trade-offs between accuracy and computational efficiency compared to existing lightweight alternatives. The source code is available at https://github.com/nttcom/TruncHierVFM.
Authors: Simone Carnemolla, Matteo Pennisi, Sarinda Samarasinghe, Giovanni Bellitto, Simone Palazzo, Daniela Giordano, Mubarak Shah, Concetto Spampinato
Abstract: Understanding and explaining the behavior of machine learning models is essential for building transparent and trustworthy AI systems. We introduce DEXTER, a data-free framework that employs diffusion models and large language models to generate global, textual explanations of visual classifiers. DEXTER operates by optimizing text prompts to synthesize class-conditional images that strongly activate a target classifier. These synthetic samples are then used to elicit detailed natural language reports that describe class-specific decision patterns and biases. Unlike prior work, DEXTER enables natural language explanation about a classifier's decision process without access to training data or ground-truth labels. We demonstrate DEXTER's flexibility across three tasks-activation maximization, slice discovery and debiasing, and bias explanation-each illustrating its ability to uncover the internal mechanisms of visual classifiers. Quantitative and qualitative evaluations, including a user study, show that DEXTER produces accurate, interpretable outputs. Experiments on ImageNet, Waterbirds, CelebA, and FairFaces confirm that DEXTER outperforms existing approaches in global model explanation and class-level bias reporting. Code is available at https://github.com/perceivelab/dexter.
Authors: Changyue Shi, Minghao Chen, Yiping Mao, Chuxiao Yang, Xinyuan Hu, Jiajun Ding, Zhou Yu
Abstract: Bridging the gap between complex human instructions and precise 3D object grounding remains a significant challenge in vision and robotics. Existing 3D segmentation methods often struggle to interpret ambiguous, reasoning-based instructions, while 2D vision-language models that excel at such reasoning lack intrinsic 3D spatial understanding. In this paper, we introduce REALM, an innovative MLLM-agent framework that enables open-world reasoning-based segmentation without requiring extensive 3D-specific post-training. We perform segmentation directly on 3D Gaussian Splatting representations, capitalizing on their ability to render photorealistic novel views that are highly suitable for MLLM comprehension. As directly feeding one or more rendered views to the MLLM can lead to high sensitivity to viewpoint selection, we propose a novel Global-to-Local Spatial Grounding strategy. Specifically, multiple global views are first fed into the MLLM agent in parallel for coarse-level localization, aggregating responses to robustly identify the target object. Then, several close-up novel views of the object are synthesized to perform fine-grained local segmentation, yielding accurate and consistent 3D masks. Extensive experiments show that REALM achieves remarkable performance in interpreting both explicit and implicit instructions across LERF, 3D-OVS, and our newly introduced REALM3D benchmarks. Furthermore, our agent framework seamlessly supports a range of 3D interaction tasks, including object removal, replacement, and style transfer, demonstrating its practical utility and versatility. Project page: https://ChangyueShi.github.io/REALM.
Authors: Hodaka Kawachi, Tomoya Nakamura, Hiroaki Santo, SaiKiran Kumar Tedla, Trevor Dalton Canham, Yasushi Yagi, Michael S. Brown
Abstract: This paper introduces a method for using LED-based environmental lighting to produce visually imperceptible watermarks for consumer cameras. Our approach optimizes an LED light source's spectral profile to be minimally visible to the human eye while remaining highly detectable by typical consumer cameras. The method jointly considers the human visual system's sensitivity to visible spectra, modern consumer camera sensors' spectral sensitivity, and narrowband LEDs' ability to generate broadband spectra perceived as "white light" (specifically, D65 illumination). To ensure imperceptibility, we employ spectral modulation rather than intensity modulation. Unlike conventional visible light communication, our approach enables watermark extraction at standard low frame rates (30-60 fps). While the information transfer rate is modest-embedding 128 bits within a 10-second video clip-this capacity is sufficient for essential metadata supporting privacy protection and content verification.
Authors: Chenxu Dang, Haiyan Liu, Jason Bao, Pei An, Xinyue Tang, PanAn, Jie Ma, Bingchuan Sun, Yan Wang
Abstract: Semantic occupancy has emerged as a powerful representation in world models for its ability to capture rich spatial semantics. However, most existing occupancy world models rely on static and fixed embeddings or grids, which inherently limit the flexibility of perception. Moreover, their ``in-place classification" over grids exhibits a potential misalignment with the dynamic and continuous nature of real scenarios. In this paper, we propose SparseWorld, a novel 4D occupancy world model that is flexible, adaptive, and efficient, powered by sparse and dynamic queries. We propose a Range-Adaptive Perception module, in which learnable queries are modulated by the ego vehicle states and enriched with temporal-spatial associations to enable extended-range perception. To effectively capture the dynamics of the scene, we design a State-Conditioned Forecasting module, which replaces classification-based forecasting with regression-guided formulation, precisely aligning the dynamic queries with the continuity of the 4D environment. In addition, We specifically devise a Temporal-Aware Self-Scheduling training strategy to enable smooth and efficient training. Extensive experiments demonstrate that SparseWorld achieves state-of-the-art performance across perception, forecasting, and planning tasks. Comprehensive visualizations and ablation studies further validate the advantages of SparseWorld in terms of flexibility, adaptability, and efficiency.
Authors: Bohan Li, Zhuang Ma, Dalong Du, Baorui Peng, Zhujin Liang, Zhenqiang Liu, Chao Ma, Yueming Jin, Hao Zhao, Wenjun Zeng, Xin Jin
Abstract: Autonomous driving world models are expected to work effectively across three core dimensions: state, action, and reward. Existing models, however, are typically restricted to limited state modalities, short video sequences, imprecise action control, and a lack of reward awareness. In this paper, we introduce OmniNWM, an omniscient panoramic navigation world model that addresses all three dimensions within a unified framework. For state, OmniNWM jointly generates panoramic videos of RGB, semantics, metric depth, and 3D occupancy. A flexible forcing strategy enables high-quality long-horizon auto-regressive generation. For action, we introduce a normalized panoramic Plucker ray-map representation that encodes input trajectories into pixel-level signals, enabling highly precise and generalizable control over panoramic video generation. Regarding reward, we move beyond learning reward functions with external image-based models: instead, we leverage the generated 3D occupancy to directly define rule-based dense rewards for driving compliance and safety. Extensive experiments demonstrate that OmniNWM achieves state-of-the-art performance in video generation, control accuracy, and long-horizon stability, while providing a reliable closed-loop evaluation framework through occupancy-grounded rewards. Project page is available at https://arlo0o.github.io/OmniNWM/.
Authors: Siyang Wu, Jack Nugent, Willow Yang, Jia Deng
Abstract: Monocular depth estimation is an important task with rapid progress, but how to evaluate it is not fully resolved, as evidenced by a lack of standardization in existing literature and a large selection of evaluation metrics whose trade-offs and behaviors are not fully understood. This paper contributes a novel, quantitative analysis of existing metrics in terms of their sensitivity to various types of perturbations of ground truth, emphasizing comparison to human judgment. Our analysis reveals that existing metrics are severely under-sensitive to curvature perturbation such as making smooth surfaces bumpy. To remedy this, we introduce a new metric based on relative surface normals, along with new depth visualization tools and a principled method to create composite metrics with better human alignment. Code and data are available at: https://github.com/princeton-vl/evalmde.
Authors: Zeyu Wang, Zilong Chen, Chenhui Gou, Feng Li, Chaorui Deng, Deyao Zhu, Kunchang Li, Weihao Yu, Haoqin Tu, Haoqi Fan, Cihang Xie
Abstract: Unified multimodal models have recently shown remarkable gains in both capability and versatility, yet most leading systems are still trained from scratch and require substantial computational resources. In this paper, we show that competitive performance can be obtained far more efficiently by strategically fusing publicly available models specialized for either generation or understanding. Our key design is to retain the original blocks while additionally interleaving multimodal self-attention blocks throughout the networks. This double fusion mechanism (1) effectively enables rich multi-modal fusion while largely preserving the original strengths of the base models, and (2) catalyzes synergistic fusion of high-level semantic representations from the understanding encoder with low-level spatial signals from the generation encoder. By training with only ~ 35B tokens, this approach achieves strong results across multiple benchmarks: 0.91 on GenEval for compositional text-to-image generation, 82.16 on DPG-Bench for complex text-to-image generation, 6.06 on GEditBench, and 3.77 on ImgEdit-Bench for image editing. By fully releasing the entire suite of code, model weights, and datasets, we hope to support future research on unified multimodal modeling.
Authors: Runxi Huang, Mingxuan Yu, Mingyu Tsoi, Xiaomin Ouyang
Abstract: Real-time multimodal inference on resource-constrained edge devices is essential for applications such as autonomous driving, human-computer interaction, and mobile health. However, prior work often overlooks the tight coupling between sensing dynamics and model execution, as well as the complex inter-modality dependencies. In this paper, we propose MMEdge, an new on-device multi-modal inference framework based on pipelined sensing and encoding. Instead of waiting for complete sensor inputs, MMEdge decomposes the entire inference process into a sequence of fine-grained sensing and encoding units, allowing computation to proceed incrementally as data arrive. MMEdge also introduces a lightweight but effective temporal aggregation module that captures rich temporal dynamics across different pipelined units to maintain accuracy performance. Such pipelined design also opens up opportunities for fine-grained cross-modal optimization and early decision-making during inference. To further enhance system performance under resource variability and input data complexity, MMEdge incorporates an adaptive multimodal configuration optimizer that dynamically selects optimal sensing and model configurations for each modality under latency constraints, and a cross-modal speculative skipping mechanism that bypasses future units of slower modalities when early predictions reach sufficient confidence. We evaluate MMEdge using two public multimodal datasets and deploy it on a real-world unmanned aerial vehicle (UAV)-based multimodal testbed. The results show that MMEdge significantly reduces end-to-end latency while maintaining high task accuracy across various system and data dynamics.
Authors: Doan-Van-Anh Ly (The Saigon International University), Thi-Thu-Hien Pham (International University, Vietnam National University HCMC), Thanh-Hai Le (The Saigon International University)
Abstract: Segmentation of liver structures in multi-phase contrast-enhanced computed tomography (CECT) plays a crucial role in computer-aided diagnosis and treatment planning for liver diseases, including tumor detection. In this study, we investigate the performance of UNet-based architectures for liver tumor segmentation, starting from the original UNet and extending to UNet3+ with various backbone networks. We evaluate ResNet, Transformer-based, and State-space (Mamba) backbones, all initialized with pretrained weights. Surprisingly, despite the advances in modern architecture, ResNet-based models consistently outperform Transformer- and Mamba-based alternatives across multiple evaluation metrics. To further improve segmentation quality, we introduce attention mechanisms into the backbone and observe that incorporating the Convolutional Block Attention Module (CBAM) yields the best performance. ResNetUNet3+ with CBAM module not only produced the best overlap metrics with a Dice score of 0.755 and IoU of 0.662, but also achieved the most precise boundary delineation, evidenced by the lowest HD95 distance of 77.911. The model's superiority was further cemented by its leading overall accuracy of 0.925 and specificity of 0.926, showcasing its robust capability in accurately identifying both lesion and healthy tissue. To further enhance interpretability, Grad-CAM visualizations were employed to highlight the region's most influential predictions, providing insights into its decision-making process. These findings demonstrate that classical ResNet architecture, when combined with modern attention modules, remain highly competitive for medical image segmentation tasks, offering a promising direction for liver tumor detection in clinical practice.
Authors: Li Wang, Yiyu Zhuang, Yanwen Wang, Xun Cao, Chuan Guo, Xinxin Zuo, Hao Zhu
Abstract: 3D human pose estimation from sketches has broad applications in computer animation and film production. Unlike traditional human pose estimation, this task presents unique challenges due to the abstract and disproportionate nature of sketches. Previous sketch-to-pose methods, constrained by the lack of large-scale sketch-3D pose annotations, primarily relied on optimization with heuristic rules-an approach that is both time-consuming and limited in generalizability. To address these challenges, we propose a novel approach leveraging a "learn from synthesis" strategy. First, a diffusion model is trained to synthesize sketch images from 2D poses projected from 3D human poses, mimicking disproportionate human structures in sketches. This process enables the creation of a synthetic dataset, SKEP-120K, consisting of 120k accurate sketch-3D pose annotation pairs across various sketch styles. Building on this synthetic dataset, we introduce an end-to-end data-driven framework for estimating human poses and shapes from diverse sketch styles. Our framework combines existing 2D pose detectors and generative diffusion priors for sketch feature extraction with a feed-forward neural network for efficient 2D pose estimation. Multiple heuristic loss functions are incorporated to guarantee geometric coherence between the derived 3D poses and the detected 2D poses while preserving accurate self-contacts. Qualitative, quantitative, and subjective evaluations collectively show that our model substantially surpasses previous ones in both estimation accuracy and speed for sketch-to-pose tasks.
Authors: Huanlin Gao, Ping Chen, Fuyuan Shi, Chao Tan, Zhaoxiang Liu, Fang Zhao, Kai Wang, Shiguo Lian
Abstract: We present LeMiCa, a training-free and efficient acceleration framework for diffusion-based video generation. While existing caching strategies primarily focus on reducing local heuristic errors, they often overlook the accumulation of global errors, leading to noticeable content degradation between accelerated and original videos. To address this issue, we formulate cache scheduling as a directed graph with error-weighted edges and introduce a Lexicographic Minimax Path Optimization strategy that explicitly bounds the worst-case path error. This approach substantially improves the consistency of global content and style across generated frames. Extensive experiments on multiple text-to-video benchmarks demonstrate that LeMiCa delivers dual improvements in both inference speed and generation quality. Notably, our method achieves a 2.9x speedup on the Latte model and reaches an LPIPS score of 0.05 on Open-Sora, outperforming prior caching techniques. Importantly, these gains come with minimal perceptual quality degradation, making LeMiCa a robust and generalizable paradigm for accelerating diffusion-based video generation. We believe this approach can serve as a strong foundation for future research on efficient and reliable video synthesis. Our code is available at :https://github.com/UnicomAI/LeMiCa
Authors: Liuzhuozheng Li, Yue Gong, Shanyuan Liu, Bo Cheng, Yuhang Ma, Liebucha Wu, Dengyang Jiang, Zanyi Wang, Dawei Leng, Yuhui Yin
Abstract: We introduce RefTON, a flux-based person-to-person virtual try-on framework that enhances garment realism through unpaired visual references. Unlike conventional approaches that rely on complex auxiliary inputs such as body parsing and warped mask or require finely designed extract branches to process various input conditions, RefTON streamlines the process by directly generating try-on results from a source image and a target garment, without the need for structural guidance or auxiliary components to handle diverse inputs. Moreover, inspired by human clothing selection behavior, RefTON leverages additional reference images (the target garment worn on different individuals) to provide powerful guidance for refining texture alignment and maintaining the garment details. To enable this capability, we built a dataset containing unpaired reference images for training. Extensive experiments on public benchmarks demonstrate that RefTON achieves competitive or superior performance compared to state-of-the-art methods, while maintaining a simple and efficient person-to-person design.
Authors: Sampriti Soor, Alik Pramanick, Jothiprakash K, Arijit Sur
Abstract: The rapid growth of deep learning has brought about powerful models that can handle various tasks, like identifying images and understanding language. However, adversarial attacks, an unnoticed alteration, can deceive models, leading to inaccurate predictions. In this paper, a generative adversarial attack method is proposed that uses the CLIP model to create highly effective and visually imperceptible adversarial perturbations. The CLIP model's ability to align text and image representation helps incorporate natural language semantics with a guided loss to generate effective adversarial examples that look identical to the original inputs. This integration allows extensive scene manipulation, creating perturbations in multi-object environments specifically designed to deceive multilabel classifiers. Our approach integrates the concentrated perturbation strategy from Saliency-based Auto-Encoder (SSAE) with the dissimilar text embeddings similar to Generative Adversarial Multi-Object Scene Attacks (GAMA), resulting in perturbations that both deceive classification models and maintain high structural similarity to the original images. The model was tested on various tasks across diverse black-box victim models. The experimental results show that our method performs competitively, achieving comparable or superior results to existing techniques, while preserving greater visual fidelity.
Authors: Zhengjie Zhang, Xiaoxie Mao, Qihao Guo, Shaoting Zhang, Qi Huang, Mu Zhou, Fang Xie, Mianxin Liu
Abstract: Background: Alzheimer's disease (AD) diagnosis heavily relies on amyloid-beta positron emission tomography (Abeta-PET), which is limited by high cost and limited accessibility. This study explores whether Abeta-PET spatial patterns can be predicted from blood-based biomarkers (BBMs) and MRI scans. Methods: We collected Abeta-PET images, T1-weighted MRI scans, and BBMs from 566 participants. A language-enhanced generative model, driven by a large language model (LLM) and multimodal information fusion, was developed to synthesize PET images. Synthesized images were evaluated for image quality, diagnostic consistency, and clinical applicability within a fully automated diagnostic pipeline. Findings: The synthetic PET images closely resemble real PET scans in both structural details (SSIM = 0.920 +/- 0.003) and regional patterns (Pearson's r = 0.955 +/- 0.007). Diagnostic outcomes using synthetic PET show high agreement with real PET-based diagnoses (accuracy = 0.80). Using synthetic PET, we developed a fully automatic AD diagnostic pipeline integrating PET synthesis and classification. The synthetic PET-based model (AUC = 0.78) outperforms T1-based (AUC = 0.68) and BBM-based (AUC = 0.73) models, while combining synthetic PET and BBMs further improved performance (AUC = 0.79). Ablation analysis supports the advantages of LLM integration and prompt engineering. Interpretation: Our language-enhanced generative model synthesizes realistic PET images, enhancing the utility of MRI and BBMs for Abeta spatial pattern assessment and improving the diagnostic workflow for Alzheimer's disease.
Authors: Tianfan Peng, Yuntao Du, Pengzhou Ji, Shijie Dong, Kailin Jiang, Mingchuan Ma, Yijun Tian, Jinhe Bi, Qian Li, Wei Du, Feng Xiao, Lizhen Cui
Abstract: Large multimodal models (LMMs) often suffer from severe inference inefficiency due to the large number of visual tokens introduced by image encoders. While recent token compression methods, such as pruning and merging, have shown promise in reducing redundancy, their evaluation remains fragmented and inconsistent. In this work, we present UniPruneBench, a unified and extensible benchmark for visual token pruning in multimodal LLMs. UniPruneBench provides standardized protocols across six ability dimensions and ten datasets, covering ten representative compression algorithms and three families of LMMs (LLaVA-v1.5, Intern-VL3, and Qwen2.5-VL). Beyond task accuracy, it incorporates system-level metrics such as runtime and prefilling latency to provide a holistic view. Our experiments uncover several key findings: (1) random pruning is a surprisingly strong baseline, (2) no single method consistently outperforms others across scenarios, (3) pruning sensitivity varies significantly across tasks, with OCR being most vulnerable, and (4) pruning ratio is the dominant factor governing performance degradation. We believe UniPruneBench will serve as a reliable foundation for future research on efficient multimodal modeling.
Authors: Shengyu Tang, Zeyuan Lu, Jiazhi Dong, Changdong Yu, Xiaoyu Wang, Yaohui Lyu, Weihao Xia
Abstract: Accurate perception of the marine environment through robust multi-object tracking (MOT) is essential for ensuring safe vessel navigation and effective maritime surveillance. However, the complicated maritime environment often causes camera motion and subsequent visual degradation, posing significant challenges to MOT. To address this challenge, we propose an efficient Dual-branch Maritime SORT (DMSORT) method for maritime MOT. The core of the framework is a parallel tracker with affine compensation, which incorporates an object detection and re-identification (ReID) branch, along with a dedicated branch for dynamic camera motion estimation. Specifically, a Reversible Columnar Detection Network (RCDN) is integrated into the detection module to leverage multi-level visual features for robust object detection. Furthermore, a lightweight Transformer-based appearance extractor (Li-TAE) is designed to capture global contextual information and generate robust appearance features. Another branch decouples platform-induced and target-intrinsic motion by constructing a projective transformation, applying platform-motion compensation within the Kalman filter, and thereby stabilizing true object trajectories. Finally, a clustering-optimized feature fusion module effectively combines motion and appearance cues to ensure identity consistency under noise, occlusion, and drift. Extensive evaluations on the Singapore Maritime Dataset demonstrate that DMSORT achieves state-of-the-art performance. Notably, DMSORT attains the fastest runtime among existing ReID-based MOT frameworks while maintaining high identity consistency and robustness to jitter and occlusion. Code is available at: https://github.com/BiscuitsLzy/DMSORT-An-efficient-parallel-maritime-multi-object-tracking-architecture-.
Authors: Claudio Giusti, Luca Guarnera, Sebastiano Battiato
Abstract: The growing sophistication of synthetic image and deepfake generation models has turned source attribution and authenticity verification into a critical challenge for modern computer vision systems. Recent studies suggest that diffusion pipelines unintentionally imprint persistent statistical traces, known as signal-leaks, within their outputs, particularly in latent representations. Building on this observation, we propose Proto-LeakNet, a signal-leak-aware and interpretable attribution framework that integrates closed-set classification with a density-based open-set evaluation on the learned embeddings, enabling analysis of unseen generators without retraining. Acting in the latent domain of diffusion models, our method re-simulates partial forward diffusion to expose residual generator-specific cues. A temporal attention encoder aggregates multi-step latent features, while a feature-weighted prototype head structures the embedding space and enables transparent attribution. Trained solely on closed data and achieving a Macro AUC of 98.13%, Proto-LeakNet learns a latent geometry that remains robust under post-processing, surpassing state-of-the-art methods, and achieves strong separability both between real images and known generators, and between known and unseen ones. The codebase will be available after acceptance.
Authors: Shirin Ermis, Cesar Aybar, Lilli Freischem, Stella Girtsou, Kyriaki-Margarita Bintsi, Emiliano Diaz Salas-Porras, Michael Eisinger, William Jones, Anna Jungbluth, Benoit Tremblay
Abstract: Accurate forecasting of tropical cyclones (TCs) remains challenging due to limited satellite observations probing TC structure and difficulties in resolving cloud properties involved in TC intensification. Recent research has demonstrated the capabilities of machine learning methods for 3D cloud reconstruction from satellite observations. However, existing approaches have been restricted to regions where TCs are uncommon, and are poorly validated for intense storms. We introduce a new framework, based on a pre-training--fine-tuning pipeline, that learns from multiple satellites with global coverage to translate 2D satellite imagery into 3D cloud maps of relevant cloud properties. We apply our model to a custom-built TC dataset to evaluate performance in the most challenging and relevant conditions. We show that we can - for the first time - create global instantaneous 3D cloud maps and accurately reconstruct the 3D structure of intense storms. Our model not only extends available satellite observations but also provides estimates when observations are missing entirely. This is crucial for advancing our understanding of TC intensification and improving forecasts.
Authors: Hongxia Li, Ying Ji, Yongxin Dong, Yuehua Feng
Abstract: Low-rank sparse regression models have been widely adopted in face recognition due to their robustness against occlusion and illumination variations. However, existing methods often suffer from insufficient feature representation and limited modeling of structured corruption across samples. To address these issues, this paper proposes a Hybrid second-order gradient Histogram based Global Low-Rank Sparse Regression (H2H-GLRSR) model. First, we propose the Histogram of Oriented Hessian (HOH) to capture second-order geometric characteristics such as curvature and ridge patterns. By fusing HOH and first-order gradient histograms, we construct a unified local descriptor, termed the Hybrid second-order gradient Histogram (H2H), which enhances structural discriminability under challenging conditions. Subsequently, the H2H features are incorporated into an extended version of the Sparse Regularized Nuclear Norm based Matrix Regression (SR\_NMR) model, where a global low-rank constraint is imposed on the residual matrix to exploit cross-sample correlations in structured noise. The resulting H2H-GLRSR model achieves superior discrimination and robustness. Experimental results on benchmark datasets demonstrate that the proposed method significantly outperforms state-of-the-art regression-based classifiers in both recognition accuracy and computational efficiency.
Authors: Yuxuan Lin, Hanjing Yan, Xuan Tong, Yang Chang, Huanzhen Wang, Ziheng Zhou, Shuyong Gao, Yan Wang, Wenqiang Zhang
Abstract: Few-shot multimodal industrial anomaly detection is a critical yet underexplored task, offering the ability to quickly adapt to complex industrial scenarios. In few-shot settings, insufficient training samples often fail to cover the diverse patterns present in test samples. This challenge can be mitigated by extracting structural commonality from a small number of training samples. In this paper, we propose a novel few-shot unsupervised multimodal industrial anomaly detection method based on structural commonality, CIF (Commonality In Few). To extract intra-class structural information, we employ hypergraphs, which are capable of modeling higher-order correlations, to capture the structural commonality within training samples, and use a memory bank to store this intra-class structural prior. Firstly, we design a semantic-aware hypergraph construction module tailored for single-semantic industrial images, from which we extract common structures to guide the construction of the memory bank. Secondly, we use a training-free hypergraph message passing module to update the visual features of test samples, reducing the distribution gap between test features and features in the memory bank. We further propose a hyperedge-guided memory search module, which utilizes structural information to assist the memory search process and reduce the false positive rate. Experimental results on the MVTec 3D-AD dataset and the Eyecandies dataset show that our method outperforms the state-of-the-art (SOTA) methods in few-shot settings. Code is available at https://github.com/Sunny5250/CIF.
Authors: Muhammad Usama, Mohammad Sadil Khan, Didier Stricker, Muhammad Zeshan Afzal
Abstract: Generating editable 3D CAD models from natural language remains challenging, as existing text-to-CAD systems either produce meshes or rely on scarce design-history data. We present NURBGen, the first framework to generate high-fidelity 3D CAD models directly from text using Non-Uniform Rational B-Splines (NURBS). To achieve this, we fine-tune a large language model (LLM) to translate free-form texts into JSON representations containing NURBS surface parameters (\textit{i.e}, control points, knot vectors, degrees, and rational weights) which can be directly converted into BRep format using Python. We further propose a hybrid representation that combines untrimmed NURBS with analytic primitives to handle trimmed surfaces and degenerate regions more robustly, while reducing token complexity. Additionally, we introduce partABC, a curated subset of the ABC dataset consisting of individual CAD components, annotated with detailed captions using an automated annotation pipeline. NURBGen demonstrates strong performance on diverse prompts, surpassing prior methods in geometric fidelity and dimensional accuracy, as confirmed by expert evaluations. Code and dataset will be released publicly.
Authors: Ardhendu Sekhar, Vasu Soni, Keshav Aske, Shivam Madnoorkar, Pranav Jeevan, Amit Sethi
Abstract: Accurate survival prediction from histopathology whole-slide images (WSIs) remains challenging due to their gigapixel resolution, strong spatial heterogeneity, and complex survival distributions. We introduce a comprehensive computational pathology framework that addresses these limitations through four complementary innovations: (1) Quantile-Gated Patch Selection for dynamically identifying prognostically relevant regions, (2) Graph-Guided Clustering to group patches by spatial and morphological similarity, (3) Hierarchical Context Attention to model both local tissue interactions and global slide-level context, and (4) an Expert-Driven Mixture of Log-Logistics module that flexibly models complex survival distributions. Across large TCGA cohorts, our method achieves state-of-the-art performance, yielding time-dependent concordance indices of 0.644 on LUAD, 0.751 on KIRC, and 0.752 on BRCA, consistently outperforming both histology-only and multimodal baselines. The framework further provides improved calibration and interpretability, advancing the use of WSIs for personalized cancer prognosis.
Authors: Xin Zuo, Chenyu Qu, Haibo Zhan, Jifeng Shen, Wankou Yang
Abstract: Recent multispectral object detection methods have primarily focused on spatial-domain feature fusion based on CNNs or Transformers, while the potential of frequency-domain feature remains underexplored. In this work, we propose a novel Spatial and Frequency Feature Reconstruction method (SFFR) method, which leverages the spatial-frequency feature representation mechanisms of the Kolmogorov-Arnold Network (KAN) to reconstruct complementary representations in both spatial and frequency domains prior to feature fusion. The core components of SFFR are the proposed Frequency Component Exchange KAN (FCEKAN) module and Multi-Scale Gaussian KAN (MSGKAN) module. The FCEKAN introduces an innovative selective frequency component exchange strategy that effectively enhances the complementarity and consistency of cross-modal features based on the frequency feature of RGB and IR images. The MSGKAN module demonstrates excellent nonlinear feature modeling capability in the spatial domain. By leveraging multi-scale Gaussian basis functions, it effectively captures the feature variations caused by scale changes at different UAV flight altitudes, significantly enhancing the model's adaptability and robustness to scale variations. It is experimentally validated that our proposed FCEKAN and MSGKAN modules are complementary and can effectively capture the frequency and spatial semantic features respectively for better feature fusion. Extensive experiments on the SeaDroneSee, DroneVehicle and DVTOD datasets demonstrate the superior performance and significant advantages of the proposed method in UAV multispectral object perception task. Code will be available at https://github.com/qchenyu1027/SFFR.
Authors: Haotian Xia, Haonan Ge, Junbo Zou, Hyun Woo Choi, Xuebin Zhang, Danny Suradja, Botao Rui, Ethan Tran, Wendy Jin, Zhen Ye, Xiyang Lin, Christopher Lai, Shengjie Zhang, Junwen Miao, Shichao Chen, Rhys Tracy, Vicente Ordonez, Weining Shen, Hanjie Chen
Abstract: Deeply understanding sports requires an intricate blend of fine-grained visual perception and rule-based reasoning - a challenge that pushes the limits of current multimodal models. To succeed, models must master three critical capabilities: perceiving nuanced visual details, applying abstract sport rule knowledge, and grounding that knowledge in specific visual evidence. Current sports benchmarks either cover single sports or lack the detailed reasoning chains and precise visual grounding needed to robustly evaluate these core capabilities in a multi-sport context. To address this gap, we introduce SportR, the first multi-sports large-scale benchmark designed to train and evaluate MLLMs on the fundamental reasoning required for sports intelligence. Our benchmark provides a dataset of 5,017 images and 2,101 videos. To enable granular evaluation, we structure our benchmark around a progressive hierarchy of question-answer (QA) pairs designed to probe reasoning at increasing depths - from simple infraction identification to complex penalty prediction. For the most advanced tasks requiring multi-step reasoning, such as determining penalties or explaining tactics, we provide 7,118 high-quality, human-authored Chain of Thought (CoT) annotations. In addition, our benchmark incorporates both image and video modalities and provides manual bounding box annotations to test visual grounding in the image part directly. Extensive experiments demonstrate the profound difficulty of our benchmark. State-of-the-art baseline models perform poorly on our most challenging tasks. While training on our data via Supervised Fine-Tuning and Reinforcement Learning improves these scores, they remain relatively low, highlighting a significant gap in current model capabilities. SportR presents a new challenge for the community, providing a critical resource to drive future research in multimodal sports reasoning.
Authors: Huiyuan Tian, Bonan Xu, Shijian Li
Abstract: While feature-based knowledge distillation has proven highly effective for compressing CNNs, these techniques unexpectedly fail when applied to Vision Transformers (ViTs), often performing worse than simple logit-based distillation. We provide the first comprehensive analysis of this phenomenon through a novel analytical framework termed as "distillation dynamics", combining frequency spectrum analysis, information entropy metrics, and activation magnitude tracking. Our investigation reveals that ViTs exhibit a distinctive U-shaped information processing pattern: initial compression followed by expansion. We identify the root cause of negative transfer in feature distillation: a fundamental representational paradigm mismatch between teacher and student models. Through frequency-domain analysis, we show that teacher models employ distributed, high-dimensional encoding strategies in later layers that smaller student models cannot replicate due to limited channel capacity. This mismatch causes late-layer feature alignment to actively harm student performance. Our findings reveal that successful knowledge transfer in ViTs requires moving beyond naive feature mimicry to methods that respect these fundamental representational constraints, providing essential theoretical guidance for designing effective ViTs compression strategies. All source code and experimental logs are provided at https://github.com/thy960112/Distillation-Dynamics.
Authors: Da-Yeong Kim, Yeong-Jun Cho
Abstract: Point cloud completion aims to recover missing geometric structures from incomplete 3D scans, which often suffer from occlusions or limited sensor viewpoints. Existing methods typically assume fixed input/output densities or rely on image-based representations, making them less suitable for real-world scenarios with variable sparsity and limited supervision. In this paper, we introduce Density-agnostic and Class-aware Network (DANCE), a novel framework that completes only the missing regions while preserving the observed geometry. DANCE generates candidate points via ray-based sampling from multiple viewpoints. A transformer decoder then refines their positions and predicts opacity scores, which determine the validity of each point for inclusion in the final surface. To incorporate semantic guidance, a lightweight classification head is trained directly on geometric features, enabling category-consistent completion without external image supervision. Extensive experiments on the PCN and MVP benchmarks show that DANCE outperforms state-of-the-art methods in accuracy and structural consistency, while remaining robust to varying input densities and noise levels.
Authors: Kaiyi Xu, Junchao Gong, Zhiwang Zhou, Zhangrui Li, Yuandong Pu, Yihao Liu, Ben Fei, Fenghua Ling, Wenlong Zhang, Lei Bai
Abstract: With the advancement of meteorological instruments, abundant data has become available. Current approaches are typically focus on single-variable, single-region tasks and primarily rely on deterministic modeling. This limits unified synthesis across variables and regions, overlooks cross-variable complementarity and often leads to over-smoothed results. To address above challenges, we introduce SynWeather, the first dataset designed for Unified Multi-region and Multi-variable Weather Observation Data Synthesis. SynWeather covers four representative regions: the Continental United States, Europe, East Asia, and Tropical Cyclone regions, as well as provides high-resolution observations of key weather variables, including Composite Radar Reflectivity, Hourly Precipitation, Visible Light, and Microwave Brightness Temperature. In addition, we introduce SynWeatherDiff, a general and probabilistic weather synthesis model built upon the Diffusion Transformer framework to address the over-smoothed problem. Experiments on the SynWeather dataset demonstrate the effectiveness of our network compared with both task-specific and general models.
Authors: Simone Ricci, Niccol\`o Biondi, Federico Pernici, Alberto Del Bimbo
Abstract: Minimizing inconsistencies across successive versions of an AI system is as crucial as reducing the overall error. In image classification, such inconsistencies manifest as negative flips, where an updated model misclassifies test samples that were previously classified correctly. This issue becomes increasingly pronounced as the number of training classes grows over time, since adding new categories reduces the margin of each class and may introduce conflicting patterns that undermine their learning process, thereby degrading performance on the original subset. To mitigate negative flips, we propose a novel approach that preserves the margins of the original model while learning an improved one. Our method encourages a larger relative margin between the previously learned and newly introduced classes by introducing an explicit margin-calibration term on the logits. However, overly constraining the logit margin for the new classes can significantly degrade their accuracy compared to a new independently trained model. To address this, we integrate a double-source focal distillation loss with the previous model and a new independently trained model, learning an appropriate decision margin from both old and new data, even under a logit margin calibration. Extensive experiments on image classification benchmarks demonstrate that our approach consistently reduces the negative flip rate with high overall accuracy.
Authors: Yifei Sun, Yuzhi He, Junhao Jia, Jinhong Wang, Ruiquan Ge, Changmiao Wang, Hongxia Xu
Abstract: Microaneurysms (MAs), the earliest pathognomonic signs of Diabetic Retinopathy (DR), present as sub-60 $\mu m$ lesions in fundus images with highly variable photometric and morphological characteristics, rendering manual screening not only labor-intensive but inherently error-prone. While diffusion-based anomaly detection has emerged as a promising approach for automated MA screening, its clinical application is hindered by three fundamental limitations. First, these models often fall prey to "identity mapping", where they inadvertently replicate the input image. Second, they struggle to distinguish MAs from other anomalies, leading to high false positives. Third, their suboptimal reconstruction of normal features hampers overall performance. To address these challenges, we propose a Wavelet Diffusion Transformer framework for MA Detection (WDT-MD), which features three key innovations: a noise-encoded image conditioning mechanism to avoid "identity mapping" by perturbing image conditions during training; pseudo-normal pattern synthesis via inpainting to introduce pixel-level supervision, enabling discrimination between MAs and other anomalies; and a wavelet diffusion Transformer architecture that combines the global modeling capability of diffusion Transformers with multi-scale wavelet analysis to enhance reconstruction of normal retinal features. Comprehensive experiments on the IDRiD and e-ophtha MA datasets demonstrate that WDT-MD outperforms state-of-the-art methods in both pixel-level and image-level MA detection. This advancement holds significant promise for improving early DR screening.
Authors: PAN Team, Jiannan Xiang, Yi Gu, Zihan Liu, Zeyu Feng, Qiyue Gao, Yiyan Hu, Benhao Huang, Guangyi Liu, Yichi Yang, Kun Zhou, Davit Abrahamyan, Arif Ahmad, Ganesh Bannur, Junrong Chen, Kimi Chen, Mingkai Deng, Ruobing Han, Xinqi Huang, Haoqiang Kang, Zheqi Liu, Enze Ma, Hector Ren, Yashowardhan Shinde, Rohan Shingre, Ramsundar Tanikella, Kaiming Tao, Dequan Yang, Xinle Yu, Cong Zeng, Binglin Zhou, Zhengzhong Liu, Zhiting Hu, Eric P. Xing
Abstract: A world model enables an intelligent agent to imagine, predict, and reason about how the world evolves in response to its actions, and accordingly to plan and strategize. While recent video generation models produce realistic visual sequences, they typically operate in the prompt-to-full-video manner without causal control, interactivity, or long-horizon consistency required for purposeful reasoning. Existing world modeling efforts, on the other hand, often focus on restricted domains (e.g., physical, game, or 3D-scene dynamics) with limited depth and controllability, and struggle to generalize across diverse environments and interaction formats. In this work, we introduce PAN, a general, interactable, and long-horizon world model that predicts future world states through high-quality video simulation conditioned on history and natural language actions. PAN employs the Generative Latent Prediction (GLP) architecture that combines an autoregressive latent dynamics backbone based on a large language model (LLM), which grounds simulation in extensive text-based knowledge and enables conditioning on language-specified actions, with a video diffusion decoder that reconstructs perceptually detailed and temporally coherent visual observations, to achieve a unification between latent space reasoning (imagination) and realizable world dynamics (reality). Trained on large-scale video-action pairs spanning diverse domains, PAN supports open-domain, action-conditioned simulation with coherent, long-term dynamics. Extensive experiments show that PAN achieves strong performance in action-conditioned world simulation, long-horizon forecasting, and simulative reasoning compared to other video generators and world models, taking a step towards general world models that enable predictive simulation of future world states for reasoning and acting.
Authors: Houzhang Fang, Shukai Guo, Qiuhuan Chen, Yi Chang, Luxin Yan
Abstract: Moving infrared small target detection (IRSTD) plays a critical role in practical applications, such as surveillance of unmanned aerial vehicles (UAVs) and UAV-based search system. Moving IRSTD still remains highly challenging due to weak target features and complex background interference. Accurate spatio-temporal feature modeling is crucial for moving target detection, typically achieved through either temporal differences or spatio-temporal (3D) convolutions. Temporal difference can explicitly leverage motion cues but exhibits limited capability in extracting spatial features, whereas 3D convolution effectively represents spatio-temporal features yet lacks explicit awareness of motion dynamics along the temporal dimension. In this paper, we propose a novel moving IRSTD network (TDCNet), which effectively extracts and enhances spatio-temporal features for accurate target detection. Specifically, we introduce a novel temporal difference convolution (TDC) re-parameterization module that comprises three parallel TDC blocks designed to capture contextual dependencies across different temporal ranges. Each TDC block fuses temporal difference and 3D convolution into a unified spatio-temporal convolution representation. This re-parameterized module can effectively capture multi-scale motion contextual features while suppressing pseudo-motion clutter in complex backgrounds, significantly improving detection performance. Moreover, we propose a TDC-guided spatio-temporal attention mechanism that performs cross-attention between the spatio-temporal features from the TDC-based backbone and a parallel 3D backbone. This mechanism models their global semantic dependencies to refine the current frame's features. Extensive experiments on IRSTD-UAV and public infrared datasets demonstrate that our TDCNet achieves state-of-the-art detection performance in moving target detection.
Authors: Ye Tian, Ling Yang, Jiongfan Yang, Anran Wang, Yu Tian, Jiani Zheng, Haochen Wang, Zhiyang Teng, Zhuochen Wang, Yinjie Wang, Yunhai Tong, Mengdi Wang, Xiangtai Li
Abstract: While thinking-aware generation aims to improve performance on complex tasks, we identify a critical failure mode where existing sequential, autoregressive approaches can paradoxically degrade performance due to error propagation. To systematically analyze this issue, we propose ParaBench, a new benchmark designed to evaluate both text and image output modalities. Our analysis using ParaBench reveals that this performance degradation is strongly correlated with poor alignment between the generated reasoning and the final image. To resolve this, we propose a parallel multimodal diffusion framework, MMaDA-Parallel, that enables continuous, bidirectional interaction between text and images throughout the entire denoising trajectory. MMaDA-Parallel is trained with supervised finetuning and then further optimized by Parallel Reinforcement Learning (ParaRL), a novel strategy that applies semantic rewards along the trajectory to enforce cross-modal consistency. Experiments validate that our model significantly improves cross-modal alignment and semantic consistency, achieving a 6.9\% improvement in Output Alignment on ParaBench compared to the state-of-the-art model, Bagel, establishing a more robust paradigm for thinking-aware image synthesis. Our code is open-sourced at https://github.com/tyfeld/MMaDA-Parallel
Authors: Felix B. Mueller, Jan F. Meier, Timo Lueddecke, Richard Vogg, Roger L. Freixanet, Valentin Hassler, Tiffany Bosshard, Elif Karakoc, William J. O'Hearn, Sofia M. Pereira, Sandro Sehner, Kaja Wierucka, Judith Burkart, Claudia Fichtel, Julia Fischer, Alexander Gail, Catherine Hobaiter, Julia Ostner, Liran Samuni, Oliver Sch\"ulke, Neda Shahidi, Erin G. Wessling, Alexander S. Ecker
Abstract: Non-human primates are our closest living relatives, and analyzing their behavior is central to research in cognition, evolution, and conservation. Computer vision could greatly aid this research, but existing methods often rely on human-centric pretrained models and focus on single datasets, which limits generalization. We address this limitation by shifting from a model-centric to a data-centric approach and introduce PriVi, a large-scale primate-centric video pretraining dataset. PriVi contains 424 hours of curated video, combining 174 hours from behavioral research across 11 settings with 250 hours of diverse web-sourced footage, assembled through a scalable data curation pipeline. We pretrain V-JEPA, a large-scale video model, on PriVi to learn primate-specific representations and evaluate it using a lightweight frozen classifier. Across four benchmark datasets, ChimpACT, BaboonLand, PanAf500, and ChimpBehave, our approach consistently outperforms prior work, including fully finetuned baselines, and scales favorably with fewer labels. These results demonstrate that primate-centric pretraining substantially improves data efficiency and generalization, making it a promising approach for low-label applications. Code, models, and the majority of the dataset will be made available.
Authors: Feng Ding, Wenhui Yi, Yunpeng Zhou, Xinan He, Hong Rao, Shu Hu
Abstract: Fairness is a core element in the trustworthy deployment of deepfake detection models, especially in the field of digital identity security. Biases in detection models toward different demographic groups, such as gender and race, may lead to systemic misjudgments, exacerbating the digital divide and social inequities. However, current fairness-enhanced detectors often improve fairness at the cost of detection accuracy. To address this challenge, we propose a dual-mechanism collaborative optimization framework. Our proposed method innovatively integrates structural fairness decoupling and global distribution alignment: decoupling channels sensitive to demographic groups at the model architectural level, and subsequently reducing the distance between the overall sample distribution and the distributions corresponding to each demographic group at the feature level. Experimental results demonstrate that, compared with other methods, our framework improves both inter-group and intra-group fairness while maintaining overall detection accuracy across domains.
Authors: Thales Bezerra, Emanoel Thyago, Kelvin Cunha, Rodrigo Abreu, F\'abio Papais, Francisco Mauro, Nat\'alia Lopes, \'Erico Medeiros, J\'essica Guido, Shirley Cruz, Paulo Borba, Tsang Ing Ren
Abstract: AI-based dermatology adoption remains limited by biased datasets, variable image quality, and limited validation. We introduce DermAI, a lightweight, smartphone-based application that enables real-time capture, annotation, and classification of skin lesions during routine consultations. Unlike prior dermoscopy-focused tools, DermAI performs on-device quality checks, and local model adaptation. The DermAI clinical dataset, encompasses a wide range of skin tones, ethinicity and source devices. In preliminary experiments, models trained on public datasets failed to generalize to our samples, while fine-tuning with local data improved performance. These results highlight the importance of standardized, diverse data collection aligned with healthcare needs and oriented to machine learning development.
Authors: Prince Mensah, Pelumi Victor Aderinto, Ibrahim Salihu Yusuf, Arnu Pretorius
Abstract: Accurate retrieval of vegetation biophysical variables from satellite imagery is crucial for ecosystem monitoring and agricultural management. In this work, we propose a physics-informed Transformer-VAE architecture to invert the PROSAIL radiative transfer model for simultaneous estimation of key canopy parameters from Sentinel-2 data. Unlike previous hybrid approaches that require real satellite images for self-supevised training. Our model is trained exclusively on simulated data, yet achieves performance on par with state-of-the-art methods that utilize real imagery. The Transformer-VAE incorporates the PROSAIL model as a differentiable physical decoder, ensuring that inferred latent variables correspond to physically plausible leaf and canopy properties. We demonstrate retrieval of leaf area index (LAI) and canopy chlorophyll content (CCC) on real-world field datasets (FRM4Veg and BelSAR) with accuracy comparable to models trained with real Sentinel-2 data. Our method requires no in-situ labels or calibration on real images, offering a cost-effective and self-supervised solution for global vegetation monitoring. The proposed approach illustrates how integrating physical models with advanced deep networks can improve the inversion of RTMs, opening new prospects for large-scale, physically-constrained remote sensing of vegetation traits.
Authors: Jiarui Zhang, Yuliang Liu, Zijun Wu, Guosheng Pang, Zhili Ye, Yupei Zhong, Junteng Ma, Tao Wei, Haiyang Xu, Weikai Chen, Zeen Wang, Qiangjun Ji, Fanxi Zhou, Qi Zhang, Yuanrui Hu, Jiahao Liu, Zhang Li, Ziyang Zhang, Qiang Liu, Xiang Bai
Abstract: Document parsing is a core task in document intelligence, supporting applications such as information extraction, retrieval-augmented generation, and automated document analysis. However, real-world documents often feature complex layouts with multi-level tables, embedded images or formulas, and cross-page structures, which remain challenging for existing OCR systems. We introduce MonkeyOCR v1.5, a unified vision-language framework that enhances both layout understanding and content recognition through a two-stage pipeline. The first stage employs a large multimodal model to jointly predict layout and reading order, leveraging visual information to ensure sequential consistency. The second stage performs localized recognition of text, formulas, and tables within detected regions, maintaining high visual fidelity while reducing error propagation. To address complex table structures, we propose a visual consistency-based reinforcement learning scheme that evaluates recognition quality via render-and-compare alignment, improving structural accuracy without manual annotations. Additionally, two specialized modules, Image-Decoupled Table Parsing and Type-Guided Table Merging, are introduced to enable reliable parsing of tables containing embedded images and reconstruction of tables crossing pages or columns. Comprehensive experiments on OmniDocBench v1.5 demonstrate that MonkeyOCR v1.5 achieves state-of-the-art performance, outperforming PPOCR-VL and MinerU 2.5 while showing exceptional robustness in visually complex document scenarios. A trial link can be found at https://github.com/Yuliang-Liu/MonkeyOCR .
Authors: Huijie Liu, Shuhao Cui, Haoxiang Cao, Shuai Ma, Kai Wu, Guoliang Kang
Abstract: Innovative visual stylization is a cornerstone of artistic creation, yet generating novel and consistent visual styles remains a significant challenge. Existing generative approaches typically rely on lengthy textual prompts, reference images, or parameter-efficient fine-tuning to guide style-aware image generation, but often struggle with style consistency, limited creativity, and complex style representations. In this paper, we affirm that a style is worth one numerical code by introducing the novel task, code-to-style image generation, which produces images with novel, consistent visual styles conditioned solely on a numerical style code. To date, this field has only been primarily explored by the industry (e.g., Midjourney), with no open-source research from the academic community. To fill this gap, we propose CoTyle, the first open-source method for this task. Specifically, we first train a discrete style codebook from a collection of images to extract style embeddings. These embeddings serve as conditions for a text-to-image diffusion model (T2I-DM) to generate stylistic images. Subsequently, we train an autoregressive style generator on the discrete style embeddings to model their distribution, allowing the synthesis of novel style embeddings. During inference, a numerical style code is mapped to a unique style embedding by the style generator, and this embedding guides the T2I-DM to generate images in the corresponding style. Unlike existing methods, our method offers unparalleled simplicity and diversity, unlocking a vast space of reproducible styles from minimal input. Extensive experiments validate that CoTyle effectively turns a numerical code into a style controller, demonstrating a style is worth one code.
Authors: Chi-Yu Chen, Rawan Abulibdeh, Arash Asgari, Leo Anthony Celi, Deirdre Goode, Hassan Hamidi, Laleh Seyyed-Kalantari, Ned McCague, Thomas Sounack, Po-Chih Kuo
Abstract: Artificial intelligence is revealing what medicine never intended to encode. Deep vision models, trained on chest X-rays, can now detect not only disease but also invisible traces of social inequality. In this study, we show that state-of-the-art architectures (DenseNet121, SwinV2-B, MedMamba) can predict a patient's health insurance type, a strong proxy for socioeconomic status, from normal chest X-rays with significant accuracy (AUC around 0.67 on MIMIC-CXR-JPG, 0.68 on CheXpert). The signal persists even when age, race, and sex are controlled for, and remains detectable when the model is trained exclusively on a single racial group. Patch-based occlusion reveals that the signal is diffuse rather than localized, embedded in the upper and mid-thoracic regions. This suggests that deep networks may be internalizing subtle traces of clinical environments, equipment differences, or care pathways; learning socioeconomic segregation itself. These findings challenge the assumption that medical images are neutral biological data. By uncovering how models perceive and exploit these hidden social signatures, this work reframes fairness in medical AI: the goal is no longer only to balance datasets or adjust thresholds, but to interrogate and disentangle the social fingerprints embedded in clinical data itself.
Authors: Quoc-Huy Trinh, Mustapha Abdullahi, Do Duy Hung Trinh, Bo Zhao, Debesh Jha
Abstract: Recent advances in multimodal large language models (MLLMs) have enabled impressive progress in vision-language understanding, yet their high computational cost limits deployment in resource-constrained scenarios such as robotic manipulation, personal assistants, and smart cameras. Most existing methods rely on Transformer-based cross-attention, whose quadratic complexity hinders efficiency. Moreover, small vision-language models often struggle to precisely capture fine-grained, task-relevant visual regions, leading to degraded performance on fine-grained reasoning tasks that limit their effectiveness in the real world. To address these issues, we introduce Viper-F1, a Hybrid State-Space Vision-Language Model that replaces attention with efficient Liquid State-Space Dynamics. To further enhance visual grounding, we propose a Token-Grid Correlation Module, which computes lightweight correlations between text tokens and image patches and modulates the state-space dynamics via FiLM conditioning. This enables the model to selectively emphasize visual regions relevant to the textual prompt while maintaining linear-time inference. Experimental results across multiple benchmarks demonstrate that Viper-F1 achieves accurate, fine-grained understanding with significantly improved efficiency.
Authors: Jitesh Chavan, Rohit Lal, Anand Kamat, Mengjia Xu
Abstract: State-space models (SSMs), Mamba in particular, are increasingly adopted for long-context sequence modeling, providing linear-time aggregation via an input-dependent, causal selective-scan operation. Along this line, recent "Mamba-for-vision" variants largely explore multiple scan orders to relax strict causality for non-sequential signals (e.g., images). Rather than preserving cross-block memory, the conventional formulation of the selective-scan operation in Mamba reinitializes each block's state-space dynamics from zero, discarding the terminal state-space representation (SSR) from the previous block. Arcee, a cross-block recurrent state chain, reuses each block's terminal state-space representation as the initial condition for the next block. Handoff across blocks is constructed as a differentiable boundary map whose Jacobian enables end-to-end gradient flow across terminal boundaries. Key to practicality, Arcee is compatible with all prior "vision-mamba" variants, parameter-free, and incurs constant, negligible cost. As a modeling perspective, we view terminal SSR as a mild directional prior induced by a causal pass over the input, rather than an estimator of the non-sequential signal itself. To quantify the impact, for unconditional generation on CelebA-HQ (256$\times$256) with Flow Matching, Arcee reduces FID$\downarrow$ from $82.81$ to $15.33$ ($5.4\times$ lower) on a single scan-order Zigzag Mamba baseline. Efficient CUDA kernels and training code will be released to support rigorous and reproducible research.
Authors: Luthira Abeykoon, Ved Patel, Gawthaman Senthilvelan, Darshan Kasundra
Abstract: Chess has experienced a large increase in viewership since the pandemic, driven largely by the accessibility of online learning platforms. However, no equivalent assistance exists for physical chess games, creating a divide between analog and digital chess experiences. This paper presents CVChess, a deep learning framework for converting chessboard images to Forsyth-Edwards Notation (FEN), which is later input into online chess engines to provide you with the best next move. Our approach employs a convolutional neural network (CNN) with residual layers to perform piece recognition from smartphone camera images. The system processes RGB images of a physical chess board through a multistep process: image preprocessing using the Hough Line Transform for edge detection, projective transform to achieve a top-down board alignment, segmentation into 64 individual squares, and piece classification into 13 classes (6 unique white pieces, 6 unique black pieces and an empty square) using the residual CNN. Residual connections help retain low-level visual features while enabling deeper feature extraction, improving accuracy and stability during training. We train and evaluate our model using the Chess Recognition Dataset (ChessReD), containing 10,800 annotated smartphone images captured under diverse lighting conditions and angles. The resulting classifications are encoded as an FEN string, which can be fed into a chess engine to generate the most optimal move
Authors: Mingzhou Jiang, Jiaying Zhou, Junde Wu, Tianyang Wang, Yueming Jin, Min Xu
Abstract: Collective intelligence from multiple medical experts consistently surpasses individual expertise in clinical diagnosis, particularly for ambiguous medical image segmentation tasks involving unclear tissue boundaries or pathological variations. The Segment Anything Model (SAM), a powerful vision foundation model originally designed for natural image segmentation, has shown remarkable potential when adapted to medical image segmentation tasks. However, existing SAM adaptation methods follow a single-expert paradigm, developing models based on individual expert annotations to predict deterministic masks. These methods systematically ignore the inherent uncertainty and variability in expert annotations, which fundamentally contradicts clinical practice, where multiple specialists provide different yet equally valid interpretations that collectively enhance diagnostic confidence. We propose an Uncertainty-aware Adapter, the first SAM adaptation framework designed to transition from single expert mindset to collective intelligence representation. Our approach integrates stochastic uncertainty sampling from a Conditional Variational Autoencoder into the adapters, enabling diverse prediction generation that captures expert knowledge distributions rather than individual expert annotations. We employ a novel position-conditioned control mechanism to integrate multi-expert knowledge, ensuring that the output distribution closely aligns with the multi-annotation distribution. Comprehensive evaluations across seven medical segmentation benchmarks have demonstrated that our collective intelligence-based adaptation achieves superior performance while maintaining computational efficiency, establishing a new adaptation framework for reliable clinical implementation.
Authors: Mona Schirmer, Dan Zhang, Eric Nalisnick
Abstract: Distribution shifts between training and test data are inevitable over the lifecycle of a deployed model, leading to performance decay. Adapting a model on test samples can help mitigate this drop in performance. However, most test-time adaptation methods have focused on synthetic corruption shifts, leaving a variety of distribution shifts underexplored. In this paper, we focus on distribution shifts that evolve gradually over time, which are common in the wild but challenging for existing methods, as we show. To address this, we propose STAD, a Bayesian filtering method that adapts a deployed model to temporal distribution shifts by learning the time-varying dynamics in the last set of hidden features. Without requiring labels, our model infers time-evolving class prototypes that act as a dynamic classification head. Through experiments on real-world temporal distribution shifts, we show that our method excels in handling small batch sizes and label shift.
Authors: Yongrok Kim, Junha Shin, Juhyun Lee, Hyunsuk Ko
Abstract: Super-Resolution (SR) is essential for displaying low-quality broadcast content on high-resolution screens. Recently, SR methods have been developed that not only increase resolution while preserving the original image information but also enhance the perceived quality. However, evaluating the quality of SR images generated from low-quality sources, such as SR-enhanced broadcast content, is challenging due to the need to consider both distortions and improvements. Additionally, assessing SR image quality without original high-quality sources presents another significant challenge. Unfortunately, there has been a dearth of research specifically addressing the Image Quality Assessment (IQA) of SR images under these conditions. In this work, we introduce a new IQA dataset for SR broadcast images in both 2K and 4K resolutions. We conducted a subjective quality evaluation to obtain Mean Opinion Score (MOS) for these SR images and performed a comprehensive human study to identify key factors influencing perceived quality. Finally, we evaluated the performance of existing IQA metrics on our dataset. This study reveals the limitations of current metrics, highlighting the need for a more robust IQA metric that better correlates with the perceived quality of SR images. The proposed dataset and the subjective evaluation platform are publicly available at https://sites.google.com/hanyang.ac.kr/ivml/datasets/sreb.
URLs: https://sites.google.com/hanyang.ac.kr/ivml/datasets/sreb.
Authors: Alexander Valverde
Abstract: Modern geometric generation methods rely heavily on deep learning methods that, while powerful, often lack interpretability and require extensive training data. This work introduces MeshCone, a convex optimization framework for mesh enhancement from partially deformed meshes that requires no training data. We formulate the problem as a second-order cone program where vertex positions are optimized to align with target geometry while enforcing smoothness through convex edge-length regularization. Our convex relaxation enables deterministic, interpretable solutions with proven convergence properties via the Splitting Conic Solver (SCS). We demonstrate robust performance across 56 diverse object categories from ShapeNet and ThreeDScans, achieving superior refinement quality compared to classical baselines while maintaining sub-second inference times. This work establishes a principled baseline demonstrating what convex optimization alone can achieve, providing mathematical guarantees and interpretability that complement data-driven approaches.
Authors: Siyuan Huang, Liliang Chen, Pengfei Zhou, Shengcong Chen, Zhengkai Jiang, Yue Hu, Yue Liao, Peng Gao, Hongsheng Li, Maoqing Yao, Guanghui Ren
Abstract: We introduce EnerVerse, a generative robotics foundation model that constructs and interprets embodied spaces. EnerVerse employs a chunk-wise autoregressive video diffusion framework to predict future embodied spaces from instructions, enhanced by a sparse context memory for long-term reasoning. To model the 3D robotics world, we adopt a multi-view video representation, providing rich perspectives to address challenges like motion ambiguity and 3D grounding. Additionally, EnerVerse-D, a data engine pipeline combining generative modeling with 4D Gaussian Splatting, forms a self-reinforcing data loop to reduce the sim-to-real gap. Leveraging these innovations, EnerVerse translates 4D world representations into physical actions via a policy head (EnerVerse-A), achieving state-of-the-art performance in both simulation and real-world tasks. For efficiency, EnerVerse-A reuses features from the first denoising step and predicts action chunks, achieving about 280 ms per 8-step action chunk on a single RTX 4090. Further video demos, dataset samples could be found in our project page.
Authors: Ping Guo, Cheng Gong, Xi Lin, Fei Liu, Zhichao Lu, Qingfu Zhang, Zhenkun Wang
Abstract: Crafting adversarial examples is crucial for evaluating and enhancing the robustness of Deep Neural Networks (DNNs), presenting a challenge equivalent to maximizing a non-differentiable 0-1 loss function. However, existing single objective methods, namely adversarial attacks focus on a surrogate loss function, do not fully harness the benefits of engaging multiple loss functions, as a result of insufficient understanding of their synergistic and conflicting nature. To overcome these limitations, we propose the Multi-Objective Set-based Attack (MOS Attack), a novel adversarial attack framework leveraging multiple loss functions and automatically uncovering their interrelations. The MOS Attack adopts a set-based multi-objective optimization strategy, enabling the incorporation of numerous loss functions without additional parameters. It also automatically mines synergistic patterns among various losses, facilitating the generation of potent adversarial attacks with fewer objectives. Extensive experiments have shown that our MOS Attack outperforms single-objective attacks. Furthermore, by harnessing the identified synergistic patterns, MOS Attack continues to show superior results with a reduced number of loss functions. Our code is available at https://github.com/pgg3/MOS-Attack.
Authors: M. Hadi Sepanj, Benyamin Ghojogh, Paul Fieguth
Abstract: Self-supervised learning has gained significant attention in contemporary applications, particularly due to the scarcity of labeled data. While existing SSL methodologies primarily address feature variance and linear correlations, they often neglect the intricate relations between samples and the nonlinear dependencies inherent in complex data--especially prevalent in high-dimensional visual data. In this paper, we introduce Correlation-Dependence Self-Supervised Learning (CDSSL), a novel framework that unifies and extends existing SSL paradigms by integrating both linear correlations and nonlinear dependencies, encapsulating sample-wise and feature-wise interactions. Our approach incorporates the Hilbert-Schmidt Independence Criterion (HSIC) to robustly capture nonlinear dependencies within a Reproducing Kernel Hilbert Space, enriching representation learning. Experimental evaluations on diverse benchmarks demonstrate the efficacy of CDSSL in improving representation quality.
Authors: Yao Du, Jiaxin Zhuang, Xiaoyu Zheng, Jing Cong, Limei Guo, Chao He, Lin Luo, Xiaomeng Li
Abstract: Histopathology image analysis is fundamental to digital pathology, with hematoxylin and eosin (H&E) staining as the gold standard for diagnostic and prognostic assessments. While H&E imaging effectively highlights cellular and tissue structures, it lacks sensitivity to birefringence and tissue anisotropy, which are crucial for assessing collagen organization, fiber alignment, and microstructural alterations--key indicators of tumor progression, fibrosis, and other pathological conditions. To bridge this gap, we construct a polarization imaging system and curate a new dataset of over 13,000 paired Polar-H&E images. Visualizations of polarization properties reveal distinctive optical signatures in pathological tissues, underscoring its diagnostic value. Building on this dataset, we propose PolarHE, a dual-modality fusion framework that integrates H&E with polarization imaging, leveraging the latter ability to enhance tissue characterization. Our approach employs a feature decomposition strategy to disentangle common and modality specific features, ensuring effective multimodal representation learning. Through comprehensive validation, our approach significantly outperforms previous methods, achieving an accuracy of 86.70% on the Chaoyang dataset and 89.06% on the MHIST dataset. These results demonstrate that polarization imaging is a powerful and underutilized modality in computational pathology, enriching feature representation and improving diagnostic accuracy. PolarHE establishes a promising direction for multimodal learning, paving the way for more interpretable and generalizable pathology models.
Authors: Chenrui Ma, Xi Xiao, Tianyang Wang, Xiao Wang, Yanning Shen
Abstract: While deep generative models have significantly advanced representation learning, they may inherit or amplify biases and fairness issues by encoding sensitive attributes alongside predictive features. Enforcing strict independence in disentanglement is often unrealistic when target and sensitive factors are naturally correlated. To address this challenge, we propose \textbf{CAD-VAE} (\textbf{C}orrelation-\textbf{A}ware \textbf{D}isentangled \textbf{VAE}), which introduces a correlated latent code to capture the information shared between the target and sensitive attributes. Given this correlated latent, our method effectively separates overlapping factors without extra domain knowledge by directly minimizing the conditional mutual information between target and sensitive codes. A relevance-driven optimization strategy refines the correlated code by efficiently capturing essential correlated features and eliminating redundancy. Extensive experiments on benchmark datasets demonstrate that CAD-VAE produces fairer representations, realistic counterfactuals, and improved fairness-aware image editing. Source code is available : https://github.com/merry7cherry/CAD-VAE
Authors: Lijie Hu, Junchi Liao, Weimin Lyu, Shaopeng Fu, Tianhao Huang, Shu Yang, Guimin Hu, Di Wang
Abstract: Backdoor attacks pose a serious threat to deep learning models by allowing adversaries to implant hidden behaviors that remain dormant on clean inputs but are maliciously triggered at inference. Existing backdoor attack methods typically rely on explicit triggers such as image patches or pixel perturbations, which makes them easier to detect and limits their applicability in complex settings. To address this limitation, we take a different perspective by analyzing backdoor attacks through the lens of concept-level reasoning, drawing on insights from interpretable AI. We show that traditional attacks can be viewed as implicitly manipulating the concepts activated within a model's latent space. This motivates a natural question: can backdoors be built by directly manipulating concepts? To answer this, we propose the Concept Confusion Attack (CCA), a novel framework that designates human-understandable concepts as internal triggers, eliminating the need for explicit input modifications. By relabeling images that strongly exhibit a chosen concept and fine-tuning on this mixed dataset, CCA teaches the model to associate the concept itself with the attacker's target label. Consequently, the presence of the concept alone is sufficient to activate the backdoor, making the attack stealthier and more resistant to existing defenses. Using CLIP as a case study, we show that CCA achieves high attack success rates while preserving clean-task accuracy and evading state-of-the-art defenses.
Authors: Yuhang Peng, Sidong Wang, Jihaoyu Yang, Shilong Li, Han Wang, Jiangtao Gong
Abstract: Most current end-to-end (E2E) autonomous driving algorithms are built on standard vehicles in structured transportation scenarios, lacking exploration of robot navigation for unstructured scenarios such as auxiliary roads, campus roads, and indoor settings. This paper investigates E2E robot navigation in unstructured road environments. First, we introduce two data collection pipelines - one for real-world robot data and another for synthetic data generated using the Isaac Sim simulator, which together produce an unstructured robotics navigation dataset -- FreeWorld Dataset. Second, we fine-tuned an efficient E2E autonomous driving model -- VAD -- using our datasets to validate the performance and adaptability of E2E autonomous driving models in these environments. Results demonstrate that fine-tuning through our datasets significantly enhances the navigation potential of E2E autonomous driving models in unstructured robotic environments. Thus, this paper presents the first dataset targeting E2E robot navigation tasks in unstructured scenarios, and provides a benchmark based on vision-based E2E autonomous driving algorithms to facilitate the development of E2E navigation technology for logistics and service robots. The project is available on Github.
Authors: Can Peng, Qianhui Men, Pramit Saha, Qianye Yang, Cheng Ouyang, J. Alison Noble
Abstract: Federated learning seeks to foster collaboration among distributed clients while preserving the privacy of their local data. Traditionally, federated learning methods assume a fixed setting in which client data and learning objectives remain constant. However, in real-world scenarios, new clients may join, and existing clients may expand the segmentation label set as task requirements evolve. In such a dynamic federated analysis setup, the conventional federated communication strategy of model aggregation per communication round is suboptimal. As new clients join, this strategy requires retraining, linearly increasing communication and computation overhead. It also imposes requirements for synchronized communication, which is difficult to achieve among distributed clients. In this paper, we propose a federated continual learning strategy that employs a one-time model aggregation at the server through multi-model distillation. This approach builds and updates the global model while eliminating the need for frequent server communication. When integrating new data streams or onboarding new clients, this approach efficiently reuses previous client models, avoiding the need to retrain the global model across the entire federation. By minimizing communication load and bypassing the need to put unchanged clients online, our approach relaxes synchronization requirements among clients, providing an efficient and scalable federated analysis framework suited for real-world applications. Using multi-class 3D abdominal CT segmentation as an application task, we demonstrate the effectiveness of the proposed approach.
Authors: Jin Zhou, Yi Zhou, Hongliang Yang, Pengfei Xu, Hui Huang
Abstract: In the field of sketch generation, raster-format trained models often produce non-stroke artifacts, while vector-format trained models typically lack a holistic understanding of sketches, leading to compromised recognizability. Moreover, existing methods struggle to extract common features from similar elements (e.g., eyes of animals) appearing at varying positions across sketches. To address these challenges, we propose StrokeFusion, a two-stage framework for vector sketch generation. It contains a dual-modal sketch feature learning network that maps strokes into a high-quality latent space. This network decomposes sketches into normalized strokes and jointly encodes stroke sequences with Unsigned Distance Function (UDF) maps, representing sketches as sets of stroke feature vectors. Building upon this representation, our framework exploits a stroke-level latent diffusion model that simultaneously adjusts stroke position, scale, and trajectory during generation. This enables high-fidelity sketch generation while supporting stroke interpolation editing. Extensive experiments on the QuickDraw dataset demonstrate that our framework outperforms state-of-the-art techniques, validating its effectiveness in preserving structural integrity and semantic features. Code and models will be made publicly available upon publication.
Authors: Haofan Wu, Yin Huang, Yuqing Wu, Qiuyu Yang, Bingfang Wang, Li Zhang, Muhammad Fahadullah Khan, Ali Zia, M. Saleh Memon, Syed Sohail Bukhari, Abdul Fattah Memon, Daizong Ji, Ya Zhang, Ghulam Mustafa, Yin Fang
Abstract: High-quality fundus images provide essential anatomical information for clinical screening and ophthalmic disease diagnosis. Yet, due to hardware limitations, operational variability, and patient compliance, fundus images often suffer from low resolution and signal-to-noise ratio. Recent years have witnessed promising progress in fundus image enhancement. However, existing works usually focus on restoring structural details or global characteristics of fundus images, lacking a unified image enhancement framework to recover comprehensive multi-scale information. Moreover, few methods pinpoint the target of image enhancement, e.g., lesions, which is crucial for medical image-based diagnosis. To address these challenges, we propose a multi-scale target-aware representation learning framework (MTRL-FIE) for efficient fundus image enhancement. Specifically, we propose a multi-scale feature encoder (MFE) that employs wavelet decomposition to embed both low-frequency structural information and high-frequency details. Next, we design a structure-preserving hierarchical decoder (SHD) to fuse multi-scale feature embeddings for real fundus image restoration. SHD integrates hierarchical fusion and group attention mechanisms to achieve adaptive feature fusion while retaining local structural smoothness. Meanwhile, a target-aware feature aggregation (TFA) module is used to enhance pathological regions and reduce artifacts. Experimental results on multiple fundus image datasets demonstrate the effectiveness and generalizability of MTRL-FIE for fundus image enhancement. Compared to state-of-the-art methods, MTRL-FIE achieves superior enhancement performance with a more lightweight architecture. Furthermore, our approach generalizes to other ophthalmic image processing tasks without supervised fine-tuning, highlighting its potential for clinical applications.
Authors: Jueqing Lu, Yuanyuan Qi, Xiaohao Yang, Shuaicheng Niu, Fucai Ke, Shujie Zhou, Wei Tan, Jionghao Lin, Wray Buntine, Hamid Rezatofighi, Lan Du
Abstract: The performance of Visio-Language Transformers drops sharply when an input modality (e.g., image) is missing, because the model is forced to make predictions using incomplete information. Existing missing-aware prompt methods help reduce this degradation, but they still rely on conventional prediction heads (e.g., a Fully-Connected layer) that compute class scores in the same way regardless of which modality is present or absent. We introduce Decoupled Prototype Learning (DPL), a new prediction head architecture that explicitly adjusts its decision process to the observed input modalities. For each class, DPL selects a set of prototypes specific to the current missing-modality cases (image-missing, text-missing, or mixed-missing). Each prototype is then decomposed into image-specific and text-specific components, enabling the head to make decisions that depend on the information actually present. This adaptive design allows DPL to handle inputs with missing modalities more effectively while remaining fully compatible with existing prompt-based frameworks. Extensive experiments on MM-IMDb, UPMC Food-101, and Hateful Memes demonstrate that DPL outperforms state-of-the-art approaches across all widely used multimodal imag-text datasets and various missing cases.
Authors: Yerin Hwang, Dongryeol Lee, Kyungmin Min, Taegwan Kang, Yong-il Kim, Kyomin Jung
Abstract: Recently, large vision-language models (LVLMs) have emerged as the preferred tools for judging text-image alignment, yet their robustness along the visual modality remains underexplored. This work is the first study to address a key research question: Can adversarial visual manipulations systematically fool LVLM judges into assigning unfairly inflated scores? We define potential image induced biases within the context of T2I evaluation and examine how these biases affect the evaluations of LVLM judges. Moreover, we introduce a novel, fine-grained, multi-domain meta-evaluation benchmark named FRAME, which is deliberately constructed to exhibit diverse score distributions. By introducing the defined biases into the benchmark, we reveal that all tested LVLM judges exhibit vulnerability across all domains, consistently inflating scores for manipulated images. Further analysis reveals that combining multiple biases amplifies their effects, and pairwise evaluations are similarly susceptible. Moreover, we observe that visual biases persist under prompt-based mitigation strategies, highlighting the vulnerability of current LVLM evaluation systems and underscoring the urgent need for more robust LVLM judges.
Authors: Taoran Zheng, Yan Yang, Xing Li, Xiang Gu, Jian Sun, Zongben Xu
Abstract: Medical image reconstruction from measurement data is a vital but challenging inverse problem. Deep learning approaches have achieved promising results, but often requires paired measurement and high-quality images, which is typically simulated through a forward model, i.e., retrospective reconstruction. However, training on simulated pairs commonly leads to performance degradation on real prospective data due to the retrospective-to-prospective gap caused by incomplete imaging knowledge in simulation. To address this challenge, this paper introduces imaging Knowledge-Informed Dynamic Optimal Transport (KIDOT), a novel dynamic optimal transport framework with optimality in the sense of preserving consistency with imaging physics in transport, that conceptualizes reconstruction as finding a dynamic transport path. KIDOT learns from unpaired data by modeling reconstruction as a continuous evolution path from measurements to images, guided by an imaging knowledge-informed cost function and transport equation. This dynamic and knowledge-aware approach enhances robustness and better leverages unpaired data while respecting acquisition physics. Theoretically, we demonstrate that KIDOT naturally generalizes dynamic optimal transport, ensuring its mathematical rationale and solution existence. Extensive experiments on MRI and CT reconstruction demonstrate KIDOT's superior performance.
Authors: Mario Leiva, Noel Ngu, Joshua Shay Kricheli, Aditya Taparia, Ransalu Senanayake, Paulo Shakarian, Nathaniel Bastian, John Corcoran, Gerardo Simari
Abstract: The deployment of pre-trained perception models in novel environments often leads to performance degradation due to distributional shifts. Although recent artificial intelligence approaches for metacognition use logical rules to characterize and filter model errors, improving precision often comes at the cost of reduced recall. This paper addresses the hypothesis that leveraging multiple pre-trained models can mitigate this recall reduction. We formulate the challenge of identifying and managing conflicting predictions from various models as a consistency-based abduction problem, building on the idea of abductive learning (ABL) but applying it to test-time instead of training. The input predictions and the learned error detection rules derived from each model are encoded in a logic program. We then seek an abductive explanation--a subset of model predictions--that maximizes prediction coverage while ensuring the rate of logical inconsistencies (derived from domain constraints) remains below a specified threshold. We propose two algorithms for this knowledge representation task: an exact method based on Integer Programming (IP) and an efficient Heuristic Search (HS). Through extensive experiments on a simulated aerial imagery dataset featuring controlled, complex distributional shifts, we demonstrate that our abduction-based framework outperforms individual models and standard ensemble baselines, achieving, for instance, average relative improvements of approximately 13.6\% in F1-score and 16.6\% in accuracy across 15 diverse test datasets when compared to the best individual model. Our results validate the use of consistency-based abduction as an effective mechanism to robustly integrate knowledge from multiple imperfect models in challenging, novel scenarios.
Authors: Ziheng Zhao, Lisong Dai, Ya Zhang, Yanfeng Wang, Weidi Xie
Abstract: Automated interpretation of CT images-particularly localizing and describing abnormal findings across multi-plane and whole-body scans-remains a significant challenge in clinical radiology. This work aims to address this challenge through four key contributions: (i) On taxonomy, we collaborate with senior radiologists to propose a comprehensive hierarchical classification system, with 404 representative abnormal findings across all body regions; (ii) On data, we contribute a dataset containing over 14.5K CT images from multiple planes and all human body regions, and meticulously provide grounding annotations for over 19K abnormalities, each linked to the detailed description and cast into the taxonomy; (iii) On model development, we propose OmniAbnorm-CT, which can automatically ground and describe abnormal findings on multi-plane and whole-body CT images based on text queries, while also allowing flexible interaction through visual prompts; (iv) On evaluation, we establish three representative tasks based on real clinical scenarios, and introduce a clinically grounded metric to assess abnormality descriptions. Through extensive experiments, we show that OmniAbnorm-CT can significantly outperform existing methods in both internal and external validations, and across all the tasks.
Authors: Weikun Peng, Jun Lv, Cewu Lu, Manolis Savva
Abstract: Articulated objects are prevalent in daily life. Interactable digital twins of such objects have numerous applications in embodied AI and robotics. Unfortunately, current methods to digitize articulated real-world objects require carefully captured data, preventing practical, scalable, and generalizable acquisition. We focus on motion analysis and part-level segmentation of an articulated object from a casually captured RGBD video shot with a hand-held camera. A casually captured video of an interaction with an articulated object is easy to obtain at scale using smartphones. However, this setting is challenging due to simultaneous object and camera motion and significant occlusions as the person interacts with the object. To tackle these challenges, we introduce iTACO: a coarse-to-fine framework that infers joint parameters and segments movable parts of the object from a dynamic RGBD video. To evaluate our method under this new setting, we build a dataset of 784 videos containing 284 objects across 11 categories that is 20$\times$ larger than available in prior work. We then compare our approach with existing methods that also take video as input. Our experiments show that iTACO outperforms existing articulated object digital twin methods on both synthetic and real casually captured RGBD videos.
Authors: Rajan Das Gupta, Md Imrul Hasan Showmick, Mushfiqur Rahman Abir, Shanjida Akter, Md. Yeasin Rahat, Md. Jakir Hossen
Abstract: Early and accurate detection of brain abnormalities, such as tumors and strokes, is essential for timely intervention and improved patient outcomes. In this study, we present a deep learning-based system capable of identifying both brain tumors and strokes from MRI images, along with their respective stages. We have executed two groundbreaking strategies involving convolutional neural networks, MobileNet V2 and ResNet-50-optimized through transfer learning to classify MRI scans into five diagnostic categories. Our dataset, aggregated and augmented from various publicly available MRI sources, was carefully curated to ensure class balance and image diversity. To enhance model generalization and prevent overfitting, we applied dropout layers and extensive data augmentation. The models achieved strong performance, with training accuracy reaching 93\% and validation accuracy up to 88\%. While ResNet-50 demonstrated slightly better results, Mobile Net V2 remains a promising option for real-time diagnosis in low resource settings due to its lightweight architecture. This research offers a practical AI-driven solution for early brain abnormality detection, with potential for clinical deployment and future enhancement through larger datasets and multi modal inputs.
Authors: Yitong Zhang, Jia Li, Liyi Cai, Ge Li
Abstract: Large Vision-Language Models (LVLMs) have achieved impressive progress across various applications but remain vulnerable to malicious queries that exploit the visual modality. Existing alignment approaches typically fail to resist malicious queries while preserving utility on benign ones effectively. To address these challenges, we propose Deep Aligned Visual Safety Prompt (DAVSP), which is built upon two key innovations. First, we introduce the Visual Safety Prompt, which appends a trainable padding region around the input image. It preserves visual features and expands the optimization space. Second, we propose Deep Alignment, a novel approach to train the visual safety prompt through supervision in the model's activation space. It enhances the inherent ability of LVLMs to perceive malicious queries, achieving deeper alignment than prior works. Extensive experiments across five benchmarks on two representative LVLMs demonstrate that DAVSP effectively resists malicious queries while preserving benign input utility. Furthermore, DAVSP exhibits great cross-model generation ability. Ablation studies further reveal that both the Visual Safety Prompt and Deep Alignment are essential components, jointly contributing to its overall effectiveness. The code is publicly available at https://github.com/zhangyitonggg/DAVSP.
Authors: Yi-Chun Chen
Abstract: We present a hierarchical knowledge graph framework for the structured semantic understanding of visual narratives, using comics as a representative domain for multimodal storytelling. The framework organizes narrative content across three levels-panel, event, and macro-event, by integrating symbolic graphs that encode semantic, spatial, and temporal relationships. At the panel level, it models visual elements such as characters, objects, and actions alongside textual components including dialogue and narration. These are systematically connected to higher-level graphs that capture narrative sequences and abstract story structures. Applied to a manually annotated subset of the Manga109 dataset, the framework supports interpretable symbolic reasoning across four representative tasks: action retrieval, dialogue tracing, character appearance mapping, and timeline reconstruction. Rather than prioritizing predictive performance, the system emphasizes transparency in narrative modeling and enables structured inference aligned with cognitive theories of event segmentation and visual storytelling. This work contributes to explainable narrative analysis and offers a foundation for authoring tools, narrative comprehension systems, and interactive media applications.
Authors: Tarun G, Naman Malpani, Gugan Thoppe, Sridharan Devarajan
Abstract: Deep neural networks are increasingly applied in automated histopathology. Yet, whole-slide images (WSIs) are often acquired at gigapixel sizes, rendering them computationally infeasible to analyze entirely at high resolution. Diagnostic labels are largely available only at the slide-level, because expert annotation of images at a finer (patch) level is both laborious and expensive. Moreover, regions with diagnostic information typically occupy only a small fraction of the WSI, making it inefficient to examine the entire slide at full resolution. Here, we propose SASHA -- Sequential Attention-based Sampling for Histopathological Analysis -- a deep reinforcement learning approach for efficient analysis of histopathological images. First, SASHA learns informative features with a lightweight hierarchical, attention-based multiple instance learning (MIL) model. Second, SASHA samples intelligently and zooms selectively into a small fraction (10-20\%) of high-resolution patches to achieve reliable diagnoses. We show that SASHA matches state-of-the-art methods that analyze the WSI fully at high resolution, albeit at a fraction of their computational and memory costs. In addition, it significantly outperforms competing, sparse sampling methods. We propose SASHA as an intelligent sampling model for medical imaging challenges that involve automated diagnosis with exceptionally large images containing sparsely informative features. Model implementation is available at: https://github.com/coglabiisc/SASHA.
Authors: Guixian Xu, Jinglai Li, Junqi Tang
Abstract: In this work, we propose Fast Equivariant Imaging (FEI), a novel unsupervised learning framework to rapidly and efficiently train deep imaging networks without ground-truth data. From the perspective of reformulating the Equivariant Imaging based optimization problem via the method of Lagrange multipliers and utilizing plug-and-play denoisers, this novel unsupervised scheme shows superior efficiency and performance compared to the vanilla Equivariant Imaging paradigm. In particular, our FEI schemes achieve an order-of-magnitude (10x) acceleration over standard EI on training U-Net for X-ray CT reconstruction and image inpainting, with improved generalization performance.
Authors: Haotian Wang, Yuzhe Weng, Jun Du, Haoran Xu, Xiaoyan Wu, Shan He, Bing Yin, Cong Liu, Jianqing Gao, Qingfeng Liu
Abstract: The introduction of diffusion models has brought significant advances to the field of audio-driven talking head generation. However, the extremely slow inference speed severely limits the practical implementation of diffusion-based talking head generation models. In this study, we propose READ, a real-time diffusion-transformer-based talking head generation framework. Our approach first learns a spatiotemporal highly compressed video latent space via a temporal VAE, significantly reducing the token count to accelerate generation. To achieve better audio-visual alignment within this compressed latent space, a pre-trained Speech Autoencoder (SpeechAE) is proposed to generate temporally compressed speech latent codes corresponding to the video latent space. These latent representations are then modeled by a carefully designed Audio-to-Video Diffusion Transformer (A2V-DiT) backbone for efficient talking head synthesis. Furthermore, to ensure temporal consistency and accelerated inference in extended generation, we propose a novel asynchronous noise scheduler (ANS) for both the training and inference processes of our framework. The ANS leverages asynchronous add-noise and asynchronous motion-guided generation in the latent space, ensuring consistency in generated video clips. Experimental results demonstrate that READ outperforms state-of-the-art methods by generating competitive talking head videos with significantly reduced runtime, achieving an optimal balance between quality and speed while maintaining robust metric stability in long-time generation.
Authors: Yuxin Cao, Wei Song, Derui Wang, Jingling Xue, Jin Song Dong
Abstract: Video Large Language Models (VideoLLMs) are increasingly deployed on numerous critical applications, where users rely on auto-generated summaries while casually skimming the video stream. We show that this interaction hides a critical safety gap: if harmful content is embedded in a video, either as full-frame inserts or as small corner patches, state-of-the-art VideoLLMs rarely mention the harmful content in the output, despite its clear visibility to human viewers. A root-cause analysis reveals three compounding design flaws: (1) insufficient temporal coverage resulting from the sparse, uniformly spaced frame sampling used by most leading VideoLLMs, (2) spatial information loss introduced by aggressive token downsampling within sampled frames, and (3) encoder-decoder disconnection, whereby visual cues are only weakly utilized during text generation. Leveraging these insights, we craft three zero-query black-box attacks, aligning with these flaws in the processing pipeline. Our large-scale evaluation across five leading VideoLLMs shows that the harmfulness omission rate exceeds 90% in most cases. Even when harmful content is clearly present in all frames, these models consistently fail to identify it. These results underscore a fundamental vulnerability in current VideoLLMs' designs and highlight the urgent need for sampling strategies, token compression, and decoding mechanisms that guarantee semantic coverage rather than speed alone.
Authors: Xiangfei Sheng, Xiaofeng Pan, Zhichao Yang, Pengfei Chen, Leida Li
Abstract: Recent years have witnessed remarkable achievements in perceptual image restoration (IR), creating an urgent demand for accurate image quality assessment (IQA), which is essential for both performance comparison and algorithm optimization. Unfortunately, the existing IQA metrics exhibit inherent weakness for IR task, particularly when distinguishing fine-grained quality differences among restored images. To address this dilemma, we contribute the first-of-its-kind fine-grained image quality assessment dataset for image restoration, termed FGRestore, comprising 18,408 restored images across six common IR tasks. Beyond conventional scalar quality scores, FGRestore was also annotated with 30,886 fine-grained pairwise preferences. Based on FGRestore, a comprehensive benchmark was conducted on the existing IQA metrics, which reveal significant inconsistencies between score-based IQA evaluations and the fine-grained restoration quality. Motivated by these findings, we further propose FGResQ, a new IQA model specifically designed for image restoration, which features both coarse-grained score regression and fine-grained quality ranking. Extensive experiments and comparisons demonstrate that FGResQ significantly outperforms state-of-the-art IQA metrics. Codes and model weights have been released in https://sxfly99.github.io/FGResQ-Homepage.
Authors: Hyun-Jic Oh, Junsik Kim, Zhiyi Shi, Yichen Wu, Yu-An Chen, Peter K. Sorger, Hanspeter Pfister, Won-Ki Jeong
Abstract: Multiplex imaging is revolutionizing pathology by enabling the simultaneous visualization of multiple biomarkers within tissue samples, providing molecular-level insights that traditional hematoxylin and eosin (H&E) staining cannot provide. However, the complexity and cost of multiplex data acquisition have hindered its widespread adoption. Additionally, most existing large repositories of H&E images lack corresponding multiplex images, limiting opportunities for multimodal analysis. To address these challenges, we leverage recent advances in latent diffusion models (LDMs), which excel at modeling complex data distributions by utilizing their powerful priors for fine-tuning to a target domain. In this paper, we introduce a novel framework for virtual multiplex staining that utilizes pretrained LDM parameters to generate multiplex images from H&E images using a conditional diffusion model. Our approach enables marker-by-marker generation by conditioning the diffusion model on each marker, while sharing the same architecture across all markers. To tackle the challenge of varying pixel value distributions across different marker stains and to improve inference speed, we fine-tune the model for single-step sampling, enhancing both color contrast fidelity and inference efficiency through pixel-level loss functions. We validate our framework on two publicly available datasets, notably demonstrating its effectiveness in generating up to 18 different marker types with improved accuracy, a substantial increase over the 2-3 marker types achieved in previous approaches. This validation highlights the potential of our framework, pioneering virtual multiplex staining. Finally, this paper bridges the gap between H&E and multiplex imaging, potentially enabling retrospective studies and large-scale analyses of existing H&E image repositories.
Authors: Hantao Zhang, Jingyang Liu, Ed Li
Abstract: We study sketch-to-diagram generation: converting rough hand sketches into precise, compositional diagrams. Diffusion models excel at photorealism but struggle with the spatial precision, alignment, and symbolic structure required for flowcharts. We introduce See it. Say it. Sorted., a training-free agentic system that couples a Vision-Language Model (VLM) with Large Language Models (LLMs) to produce editable Scalable Vector Graphics (SVG) programs. The system runs an iterative loop in which a Critic VLM proposes a small set of qualitative, relational edits; multiple candidate LLMs synthesize SVG updates with diverse strategies (conservative->aggressive, alternative, focused); and a Judge VLM selects the best candidate, ensuring stable improvement. This design prioritizes qualitative reasoning over brittle numerical estimates, preserves global constraints (e.g., alignment, connectivity), and naturally supports human-in-the-loop corrections. On 10 sketches derived from flowcharts in published papers, our method more faithfully reconstructs layout and structure than two frontier closed-source image generation LLMs (GPT-5 and Gemini-2.5-Pro), accurately composing primitives (e.g., multi-headed arrows) without inserting unwanted text. Because outputs are programmatic SVGs, the approach is readily extensible to presentation tools (e.g., PowerPoint) via APIs and can be specialized with improved prompts and task-specific tools. The codebase is open-sourced at https://github.com/hantaoZhangrichard/see_it_say_it_sorted.git.
URLs: https://github.com/hantaoZhangrichard/see_it_say_it_sorted.git.
Authors: U\u{g}urcan Aky\"uz, Deniz Katircioglu-\"Ozt\"urk, Emre K. S\"usl\"u, Burhan Kele\c{s}, Mete C. Kaya, Gamze Durhan, Meltem G. Akp{\i}nar, Figen B. Demirkaz{\i}k, G\"ozde B. Akar
Abstract: Numerous deep learning-based solutions have been developed for the automatic recognition of breast cancer using mammography images. However, their performance often declines when applied to data from different domains, primarily due to domain shift - the variation in data distributions between source and target domains. This performance drop limits the safe and equitable deployment of AI in real-world clinical settings. In this study, we present DoSReMC (Domain Shift Resilient Mammography Classification), a batch normalization (BN) adaptation framework designed to enhance cross-domain generalization without retraining the entire model. Using three large-scale full-field digital mammography (FFDM) datasets - including HCTP, a newly introduced, pathologically confirmed in-house dataset - we conduct a systematic cross-domain evaluation with convolutional neural networks (CNNs). Our results demonstrate that BN layers are a primary source of domain dependence: they perform effectively when training and testing occur within the same domain, and they significantly impair model generalization under domain shift. DoSReMC addresses this limitation by fine-tuning only the BN and fully connected (FC) layers, while preserving pretrained convolutional filters. We further integrate this targeted adaptation with an adversarial training scheme, yielding additional improvements in cross-domain generalizability while reducing the computational cost of model training. DoSReMC can be readily incorporated into existing AI pipelines and applied across diverse clinical environments, providing a practical pathway toward more robust and generalizable mammography classification systems.
Authors: Yuxuan Zhou, Xingxing Li, Shengyu Li, Zhuohao Yan, Chunxi Xia, Shaoquan Feng
Abstract: Visual SLAM is a cornerstone technique in robotics, autonomous driving and extended reality (XR), yet classical systems often struggle with low-texture environments, scale ambiguity, and degraded performance under challenging visual conditions. Recent advancements in feed-forward neural network-based pointmap regression have demonstrated the potential to recover high-fidelity 3D scene geometry directly from images, leveraging learned spatial priors to overcome limitations of traditional multi-view geometry methods. However, the widely validated advantages of probabilistic multi-sensor information fusion are often discarded in these pipelines. In this work, we propose MASt3R-Fusion,a multi-sensor-assisted visual SLAM framework that tightly integrates feed-forward pointmap regression with complementary sensor information, including inertial measurements and GNSS data. The system introduces Sim(3)-based visualalignment constraints (in the Hessian form) into a universal metric-scale SE(3) factor graph for effective information fusion. A hierarchical factor graph design is developed, which allows both real-time sliding-window optimization and global optimization with aggressive loop closures, enabling real-time pose tracking, metric-scale structure perception and globally consistent mapping. We evaluate our approach on both public benchmarks and self-collected datasets, demonstrating substantial improvements in accuracy and robustness over existing visual-centered multi-sensor SLAM systems. The code will be released open-source to support reproducibility and further research (https://github.com/GREAT-WHU/MASt3R-Fusion).
Authors: Dan Jacobellis, Mateen Ulhaq, Fabien Racap\'e, Hyomin Choi, Neeraja J. Yadwadkar
Abstract: Video comprises the vast majority of bits that are generated daily, and is the primary signal driving current innovations in robotics, remote sensing, and wearable technology. Yet, the most powerful video understanding models are too expensive for the resource-constrained platforms used in these applications. One approach is to offload inference to the cloud; this gives access to GPUs capable of processing high-resolution videos in real time. But even with reliable, high-bandwidth communication channels, the combined latency of video encoding, model inference, and round-trip communication prohibits use for certain real-time applications. The alternative is to use fully local inference; but this places extreme constraints on computational and power costs, requiring smaller models and lower resolution, leading to degraded accuracy. To address these challenges, we propose Dedelayed, a real-time inference system that divides computation between a remote model operating on delayed video frames and a local model with access to the current frame. The remote model is trained to make predictions on anticipated future frames, which the local model incorporates into its prediction for the current frame. The local and remote models are jointly optimized with an autoencoder that limits the transmission bitrate required by the available downlink communication channel. We evaluate Dedelayed on the task of real-time streaming video segmentation using the BDD100k driving dataset. For a round trip delay of 100 ms, Dedelayed improves performance by 6.4 mIoU compared to fully local inference and 9.8 mIoU compared to remote inference -- an equivalent improvement to using a model ten times larger.
Authors: Yue Xun, Jiaxing Xu, Wenbo Gao, Chen Yang, Shujun Wang
Abstract: Resting-state fMRI has become a valuable tool for classifying brain disorders and constructing brain functional connectivity networks by tracking BOLD signals across brain regions. However, existing mod els largely neglect the multi-frequency nature of neuronal oscillations, treating BOLD signals as monolithic time series. This overlooks the cru cial fact that neurological disorders often manifest as disruptions within specific frequency bands, limiting diagnostic sensitivity and specificity. While some methods have attempted to incorporate frequency informa tion, they often rely on predefined frequency bands, which may not be optimal for capturing individual variability or disease-specific alterations. To address this, we propose a novel framework featuring Adaptive Cas cade Decomposition to learn task-relevant frequency sub-bands for each brain region and Frequency-Coupled Connectivity Learning to capture both intra- and nuanced cross-band interactions in a unified functional network. This unified network informs a novel message-passing mecha nism within our Unified-GCN, generating refined node representations for diagnostic prediction. Experimental results on the ADNI and ABIDE datasets demonstrate superior performance over existing methods. The code is available at https://github.com/XXYY20221234/Ada-FCN.
Authors: Yaxin Xiao, Qingqing Ye, Zi Liang, Haoyang Li, RongHua Li, Huadi Zheng, Haibo Hu
Abstract: Machine learning models constitute valuable intellectual property, yet remain vulnerable to model extraction attacks (MEA), where adversaries replicate their functionality through black-box queries. Model watermarking counters MEAs by embedding forensic markers for ownership verification. Current black-box watermarks prioritize MEA survival through representation entanglement, yet inadequately explore resilience against sequential MEAs and removal attacks. Our study reveals that this risk is underestimated because existing removal methods are weakened by entanglement. To address this gap, we propose Watermark Removal attacK (WRK), which circumvents entanglement constraints by exploiting decision boundaries shaped by prevailing sample-level watermark artifacts. WRK effectively reduces watermark success rates by at least 88.79% across existing watermarking benchmarks. For robust protection, we propose Class-Feature Watermarks (CFW), which improve resilience by leveraging class-level artifacts. CFW constructs a synthetic class using out-of-domain samples, eliminating vulnerable decision boundaries between original domain samples and their artifact-modified counterparts (watermark samples). CFW concurrently optimizes both MEA transferability and post-MEA stability. Experiments across multiple domains show that CFW consistently outperforms prior methods in resilience, maintaining a watermark success rate of at least 70.15% in extracted models even under the combined MEA and WRK distortion, while preserving the utility of protected models.
Authors: Keshara Weerasinghe, Xueren Ge, Tessa Heick, Lahiru Nuwan Wijayasingha, Anthony Cortez, Abhishek Satpathy, John Stankovic, Homa Alemzadeh
Abstract: Emergency Medical Services (EMS) are critical to patient survival in emergencies, but first responders often face intense cognitive demands in high-stakes situations. AI cognitive assistants, acting as virtual partners, have the potential to ease this burden by supporting real-time data collection and decision making. In pursuit of this vision, we introduce EgoEMS, the first end-to-end, high-fidelity, multimodal, multiperson dataset capturing over 20 hours of realistic, procedural EMS activities from an egocentric view in 233 simulated emergency scenarios performed by 62 participants, including 46 EMS professionals. Developed in collaboration with EMS experts and aligned with national standards, EgoEMS is captured using an open-source, low-cost, and replicable data collection system and is annotated with keysteps, timestamped audio transcripts with speaker diarization, action quality metrics, and bounding boxes with segmentation masks. Emphasizing realism, the dataset includes responder-patient interactions reflecting real-world emergency dynamics. We also present a suite of benchmarks for real-time multimodal keystep recognition and action quality estimation, essential for developing AI support tools for EMS. We hope EgoEMS inspires the research community to push the boundaries of intelligent EMS systems and ultimately contribute to improved patient outcomes.
Authors: Xuanyu Tian, Lixuan Chen, Qing Wu, Xiao Wang, Jie Feng, Yuyao Zhang, Hongjiang Wei
Abstract: Cardiac magnetic resonance (CMR) imaging is widely used to characterize cardiac morphology and function. To accelerate CMR imaging, various methods have been proposed to recover high-quality spatiotemporal CMR images from highly undersampled k-t space data. However, current CMR reconstruction techniques either fail to achieve satisfactory image quality or are restricted by the scarcity of ground truth data, leading to limited applicability in clinical scenarios. In this work, we proposed MoCo-INR, a new unsupervised method that integrates implicit neural representations (INR) with the conventional motion-compensated (MoCo) framework. Using explicit motion modeling and the continuous prior of INRs, MoCo-INR can produce accurate cardiac motion decomposition and high-quality CMR reconstruction. Furthermore, we introduce a new INR network architecture tailored to the CMR problem, which significantly stabilizes model optimization. Experiments on retrospective (simulated) datasets demonstrate the superiority of MoCo-INR over state-of-the-art methods, achieving fast convergence and fine-detailed reconstructions at ultra-high acceleration factors (e.g., 20x in VISTA sampling). Additionally, evaluations on prospective (real-acquired) free-breathing CMR scans highlight the clinical practicality of MoCo-INR for real-time imaging. Several ablation studies further confirm the effectiveness of the critical components of MoCo-INR.