Authors: Diego Belzarena (UDELAR, CB), Seginus Mowlavi (CB), Paula Casariego Casti\~neira (ROMA TRE), Alejandra Ulla Lorenzo (USC), Gregory Randall (UDELAR), Jean-Michel Morel (LU - Hong Kong)
Abstract: We propose a statistical methodology that quantifies the similarity of typefaces between printed historical books. This provides a tool that accelerates philological analysis. Using character prototypes derived from clustering and aligning automatically extracted character images, the method defines a typeface distance between any two books. To produce actionable outputs, we develop an a contrario statistical framework to interpret the significance of the computed typeface distances. We apply the method to the philological study of 17 th -century Spanish printed theatre chapbooks in a quantity that exceeds the capabilities of systematic visual inspection by human experts. Our method enables the automatic comparison of Roman and Italic types extracted from different books. After validation by human experts, our method has led to new printer attributions being discovered, and former printer attributions being revised. This success strongly suggests that our method has the potential to enable digital bibliography on a larger scale than was previously possible.
Authors: Kaiyu Li, Zepeng Xin, Zixuan Jiang, Jing Fu, Lanxuan Xue, Lingyu Zhang, Xiangyong Cao
Abstract: Open-vocabulary Earth observation (EO) aims to localize geospatial concepts specified in natural language rather than a fixed label set. Existing benchmarks, however, usually cover narrow category vocabularies or limited query forms. To fill this gap, we introduce OVEarth-Bench, which extends existing evaluation in two directions: category breadth, through broad hierarchical category coverage with positive and negative expressions, and query diversity, through vocabulary, referring, and reasoning queries. The benchmark supports mask and box localization under a unified zero-shot protocol. We evaluate a broad set of general and EO-specific methods. The evaluation reveals that: (1) the performance of current methods remains limited, while broader category coverage yields more stable model rankings; (2) MLLM-based methods achieve the strongest overall performance; and (3) EO-specific methods generally underperform general models and rarely match the strongest methods. These findings provide guidance for future open-vocabulary EO method design and highlight the importance of developing more realistic, diverse, high-quality, and large-scale benchmarks for reliable evaluation. Our data and evaluation package are released at https://earth-insights.github.io/OVEarth-bench.
Authors: Jonas Grebe, Hossein Shakibania, Tobias Braun, Marcus Rohrbach, Anna Rohrbach
Abstract: The rise of powerful, accessible image-editing models such as FLUX.2 has brought high-fidelity editing within broad reach. Their capabilities now extend beyond localized modifications to extracting and recontextualizing objects and identities in entirely new scenes. By allowing prompt and generation tokens to attend directly to reference-image tokens, modern models blur the boundary between conventional editing and text-to-image synthesis. This expanded generative freedom also broadens the space of potential misuse, as harmful transformations are no longer confined to a predictable set of localized edits. Existing anti-edit defenses are designed to disrupt the semantic bottleneck of the reference-image encoding in legacy diffusion pipelines. However, newer editors distill reference information through joint-attention blocks, thereby often circumventing these protections. We therefore introduce VETO, a subtle anti-edit cloak that disrupts this inner mechanism through which modern models read the source image. Additionally, as existing editing benchmarks leave comprehensive recontextualizations largely untested, we introduce VetoBench, which evaluates defenses not only on conventional localized edits but also on broader contextual shifts. Across two contemporary editing models and three benchmarks, VETO consistently outperforms existing defenses while providing a stronger protection-fidelity trade-off.
Authors: Junhao Chen, Jingjia Mao, Dayong Li, Chenghai Li, Saining Zhang, Zhihao Li, Hao Zhao, Yufei Wang, Ruqi Huang
Abstract: Live2D is the dominant 2D character-animation format for anime characters and virtual avatars, representing each character as a stack of RGBA layers driven by per-layer mesh deformation. Despite its wide use in virtual streaming, mobile games, and interactive characters, authoring a Live2D model still demands weeks of manual layer separation, occlusion completion, mesh placement, and keyframing, and no prior generative method produces such a structured asset end-to-end. We present the first system that, from a single illustration, generates all the structured information a Live2D runtime consumes: ordered RGBA layers, a deformation mesh per layer, and the parameter-driven keypose vertex offsets that make the character move. Stage 1 casts layered decomposition as a layered diffusion process under a Live2D-aware organ-level taxonomy, producing an ordered RGBA stack with hidden-region completion. Stage 2 builds a content-conforming triangle mesh for each layer from its alpha channel alone, then predicts the keypose displacement field of all layers jointly: every vertex of every layer is one token, self-attention spans layer boundaries, and each displacement is factorised into a bounded direction and a log-magnitude. Joint rather than independent prediction is what makes the result a coherent character instead of separately plausible parts, and is our largest gain; scaling the network 112x yields none. On 50 held-out characters, under true generation with no teacher forcing, Stage 2 attains a per-vertex direction cosine of 0.768 (median 0.828). Because a layer's mesh derives from its alpha channel, a clothing layer can be re-textured from a natural-language instruction while the mesh and predicted animation are reused byte-for-byte. We further contribute Live2D-Bench, the first standardized benchmark for the task, and an 8,884-model Live2D corpus with layer and animation supervision.
Authors: Dillan Imans, Phuoc-Nguyen Bui, Duc-Tai Le, Hyunseung Choo
Abstract: Cross-modal knowledge distillation can transfer diagnostic knowledge from a strong but costly teacher modality to a cheaper and more deployable student modality. In medical image analysis, however, the two modalities are often unpaired: they are collected from different patient cohorts and occupy geometrically incompatible feature spaces. This makes instance-level distillation invalid and direct feature matching unreliable. To address these challenges, we propose Shared Semantic Codebook Distillation (SSCD), which compares teacher and student representations through a shared discrete codebook. Each image is represented as a distribution over a common, modality-agnostic vocabulary, and knowledge is transferred by aligning these distributions across modalities, both globally and class-conditionally, without requiring paired samples or directly comparable raw features. The codebook is evolved online by exponential moving average and kept diverse through entropy regularization and dead-code restart. At inference, all teacher-side and codebook modules are discarded, leaving only the student encoder and classifier. On two heterogeneous unpaired settings, OCT-to-fundus retinal disease classification and CT-to-chest-X-ray pneumonia classification, SSCD improves the student from 64.5 to 70.2 macro-F1 and from 73.8 to 76.3 macro-F1, respectively, outperforming all evaluated distillation baselines on both settings. Code and pretrained models are available at https://github.com/DillanImans/SSCD-unpaired-distillation
URLs: https://github.com/DillanImans/SSCD-unpaired-distillation
Authors: Xiaohan Li, Xinyu Liu, Chang Liu, Sum Wing Au Yeung, Jun Liu, Yixuan Yuan, Hui Chen
Abstract: Accurate evaluation of multimodal large language models (MLLMs) in dental panoramic radiography (orthopantomogram, OPG) is limited by the lack of fine-grained, clinically reliable benchmarks that reflect expert interpretation. This work introduces PanDent, a large-scale, clinically grounded OPG benchmark built upon fine-grained, expert-validated tooth-level annotations. The dataset comprises 9,524 high-quality OPGs, each associated with comprehensive structured annotations produced by experienced dentists and further validated by an oral and maxillofacial radiologist, providing clinically reliable supervision for tooth-level diagnosis and reasoning. Clinically consistent radiology reports are constructed from expert-validated findings using clinician-defined reporting logic, establishing explicit correspondence between structured clinical evidence and free-text descriptions. This design enables evaluation of whether MLLMs generate reports that are not only linguistically coherent but also clinically consistent with expert-validated tooth-level findings. Experiments are conducted on diverse MLLMs, including state-of-the-art (SOTA) proprietary models, general-domain open-source models, and medical-specific models. Results show that current MLLMs can generate fluent reports, yet fail to produce clinically consistent descriptions, exhibiting substantial errors in fine-grained localization and tooth-level diagnosis. Fine-tuning on PanDent significantly improves structure-language consistency, substantially enhancing visual localization accuracy and diagnostic correctness, and bringing model outputs closer to expert dental interpretation. These results establish PanDent as a rigorous benchmark for evaluating tooth-level clinical reasoning in MLLMs and a valuable resource for clinically grounded dental AI.
Authors: Haodong Li, Tianfei Ren, Xiaoxiao Ma, Chunmei Qing, Zhen Fang, Sipeng He, Ziyu Guo, Haoyu Wu, Juanxi Tian, Yihang Zou, Ruichuan An, Dongzhi Jiang, Boxue Yang, Ji Xie, Xu Huang, Wenhao Yan, Jialv Zou, Zhengrong Yue, Yaxin Luo, Xiaotong Li, Yuzhu Wang, Junyan Ye, Jinjing Zhao, Zehui Chen, Lin Chen, Renye Yan, Feng Zhao, Pheng-Ann Heng
Abstract: Text-to-video models have achieved remarkable visual quality, yet they still struggle to generate physically consistent dynamics because the temporal evolution of a scene must be inferred implicitly from a highly compressed text prompt. Existing chain-of-thought approaches introduce intermediate plans or visual states, but these representations are typically non-executable or temporally sparse, limiting their ability to instantiate and control the complete spatiotemporal process. To address this limitation, we introduce VideoCoCo, an agentic dual-engine framework in which executable Blender code serves as a process-level chain of thought. Given a text prompt, a coding agent synthesizes a Blender program that explicitly specifies the scene and its temporal evolution. The executable simulation engine runs the program to produce a deterministic spatiotemporal draft, which is subsequently transformed into a photorealistic video by a generative video engine through draft-conditioned editing. This decomposition separates process-level reasoning from high-fidelity visual realization. To adapt the video editor to simulated drafts, we construct VideoCoCo-3K, a curated dataset of draft-instruction-target triplets. VideoCoCo improves the OmniWeaving baseline from 0.475 to 0.558 on PhyGenBench and from 52.18 to 77.88 on VBench-2.0, achieving the best average score on both benchmarks. These results demonstrate that executable code provides an effective, controllable, and inspectable intermediate representation for physically consistent video generation.
Authors: Mengqi He, Jing Zhang
Abstract: Semantic segmentation models are vulnerable to transferable adversarial perturbations, yet evaluating transfer attacks on dense prediction models can be computationally expensive. Existing ensemble attacks often rely on multiple surrogate models, increasing the computation cost, even harder for segmentation. This paper studies an efficient single-source alternative for transferable attacks on semantic segmentation. We formulate transferable attack composition as a chained computation over differentiable attack components, allowing the expensive source-model gradient computation to be shared. To reduce the update instability introduced by chained composition, we further use an integrated-gradient-style path-averaged direction as an empirical stabilization heuristic. Experiments on Pascal VOC and Cityscapes evaluate the resulting transferability efficiency trade-off across CNN- and transformer-based segmentation models. IGME achieves competitive transferability compared with single-source baselines and favorable runtime compared with model-ensemble attacks, while requiring access to only one source model.
Authors: Dexuan Ding, Yuankai Qi, Luping Zhou, Jian Yang, Quan Z. Sheng, Ming-Hsuan Yang
Abstract: Predicting future structural MRI of a brain is challenging because longitudinal changes are often subtle and confined to specific anatomical regions, while most subject-specific brain structure remains stable over time. An effective model should therefore preserve global brain structural consistency while remaining sensitive to fine-grained disease progression. Existing latent-space-based methods improve computational efficiency, but suffer from information loss during their compression-reconstruction procedure. In contrast, direct voxel-space methods avoid latent reconstruction but commonly use a unified prediction pathway to model brain structure and progression-related changes. Subtle local changes may therefore be overshadowed by the dominant stable brain structure. To address these challenges, we propose ProgFormer, a hierarchical voxel-space Diffusion Transformer for longitudinal brain MRI prediction. ProgFormer uses a coarse pathway to perform the primary volumetric prediction from 3D patch tokens. This pathway models overall brain structure and longitudinal context. The fine pathway then uses the coarse representations as spatio-temporal grounding for voxel-level refinement within individual patches. The two pathways jointly estimate a velocity field directly in voxel space through conditional flow matching, enabling end-to-end prediction without a separately learned image autoencoder. The predicted future scan is then generated from Gaussian noise by integrating the estimated velocity field over a sequence of Euler steps. Extensive experimental results on three widely used benchmarks, ADNI, AIBL, and OASIS, under both pairwise and trajectory settings demonstrate favourable performance compared against several state-of-the-art methods.
Authors: Mingi Kim, Yongjun Kim, Hyungki Kim
Abstract: Recovering Parametric CAD sequences from raster-format 2D Computer-Aided Design (CAD) drawings accumulated prior to digital transformation is important for part reproduction and manufacturing process automation. However, existing studies either process only vector drawings or are limited to specific domains, and fail to explicitly connect dimensional annotations to geometric information, limiting their use of dimensional information for 3D Parametric CAD sequences recovery. We propose Drawing-Recode, a framework that generates Parametric CAD sequences as CAD code from raster 2D CAD drawings. Drawing-Recode extracts geometric features via an image encoder and recognizes annotations through a separate text recognition module, then explicitly grounds annotations to geometric information using cross-attention and our proposed Annotation Grounding Loss (AGL). The resulting features are fed into a Large Language Model (LLM) to generate CAD code in the Structured Parametric CAD Code (SPCC) format. Experiments show that Drawing-Recode outperforms existing baselines and remains robust on scanned drawings resembling industrial conditions. We expect Drawing-Recode contributes to digitizing raster 2D CAD drawings in industrial settings and to part reproduction and manufacturing automation.
Authors: Site Li, Jianyi Hao, Xiaofeng Liu
Abstract: Multi-image medical VQA is not merely a prompt-length problem; it is a fundamental challenge of agentic decision-making. Medical vision-language agents must aggregate evidence across ordered images, remain robust to answer-order perturbations, and avoid overfitting to noisy search-time feedback. We study MedFrameQA through a controlled comparison of five inference-time agentic strategies, optimized using the same high-budget ShinkaEvolve configuration and evaluated on a reproducible internal frozen split (1,331 evolution, 665 holdout, 855 final test). Across five independent repeated runs, the strongest method emerges as the simplest robust aggregator: the \textbf{order-vote} policy achieves $57.89 \pm 0.65\%$ final-test accuracy, significantly outperforming the fixed baseline ($52.73 \pm 0.42\%$) and the more complex, albeit brittle, order-rerank variant ($55.79 \pm 0.43\%$). Paired bootstrap analysis confirms these significant gains. Extending the evolutionary search budget from 50 to 100 generations yields no generalization benefit: while holdout performance marginally increases, final-test accuracy drops from $57.89\%$ to $56.02\%$. Our findings suggest that for multi-image medical reasoning, defining the correct agentic decision rule is substantially more impactful than expanding the optimization search budget.
Authors: Site Li, Jianyi Hao, Xiaofeng Liu
Abstract: Multi-frame medical VQA appears to reward increasingly complex adaptation: controller-style inference, localization-aware reranking, static hard-negative mixing, and staged continuation all appear plausible from first principles. We test a simpler competing hypothesis on MedFrameQA: methods that remain tightly aligned with the benchmark's final answer objective should be the strongest \emph{robust} adaptation family once evaluation is controlled across fixed splits, matched budgets, repeated seeds, and calibration. We compare controller-based methods, scaffold evolution, static mixed supervision, continuation-heavy variants, and direct answer-only supervised fine-tuning (SFT). The strongest robust family is direct decoder-only answer SFT on MedGemma-1.5-4B. Empirically, this family yields substantial improvements in held-out report accuracy over frozen baselines while remaining remarkably stable across repeated seeds and matched controls, ensuring our claims reflect true family-level robustness rather than an isolated hyperparameter peak. Furthermore, post-hoc calibration effectively repairs confidence estimation without compromising accuracy, and the core approach transfers consistently to secondary backbones like Qwen2.5-VL-3B. The main result is therefore not that a complex auxiliary mechanism wins, but that objective-aligned direct answer SFT is the strongest robust adaptation family we found for MedFrameQA. By establishing this strong, minimalist baseline, we hope to redirect community focus toward fundamentally robust optimization rather than architectural complexity.
Authors: Dekun Yuan, Zhongwei Li, Zheng Qiao, Jie Zhang
Abstract: As primary consumers in the marine food chain, zooplankton play a crucial role in maintaining marine ecological balance. However, the Segment Anything Model (SAM) exhibits limited performance in microscopic image instance segmentation due to its lack of zooplankton-specific domain knowledge. To address these challenges, we propose a novel instance segmentation model based on SAM and wavelet transform (ZMIS-SAM), effectively tackling issues such as inaccurate classification, discontinuous segmentation of slender appendages, and incomplete boundary segmentation. Our framework incorporates three core innovations: ZM-ViT enhances SAM's capability to model zooplankton morphology and image intensity distributions through two lightweight adapters, the Neighboring Feature Aggregation Module (NFAM) improves continuous segmentation of semi-transparent slender appendages by integrating general-purpose and domain-specific features, and the Wavelet-based Multi-scale Multi-directional Feature Enhancement (WM2FE) module effectively recovers high-frequency details to refine boundary segmentation completeness. Extensive experiments demonstrate that ZMIS-SAM achieves state-of-the-art instance segmentation performance on the zooplankton dataset and exhibits strong generalization capability across multiple public cross-domain datasets.
Authors: Jinfan Zhou, Richard Liu, Itai Lang, Rana Hanocka
Abstract: We present MeshFM, an efficient feedforward framework for extracting rich features from 3D inputs. Our method distills 2D features from visual foundation models into 3D. We train a feedforward network to directly predict 3D features without requiring optimization during inference. The approach utilizes a two-stage training strategy. First, we optimize a feature field in 3D using only 2D feature supervision. Second, we train a network to regress this feature field. The entire procedure requires no 3D annotation, instead relying on the powerful information in 2D foundation models. We demonstrate that our learned features can be immediately applied to downstream tasks, including part segmentation, dense correspondence, and mesh deformation. Extensive experiments show that MeshFM, trained solely with 2D supervision, performs on par with methods trained explicitly with 3D supervision, even without task-specific fine-tuning. Moreover, our model is trained to be robust to extreme rotations of the input objects. Project page: https://threedle.github.io/MeshFM/
Authors: Jiajia Lin, Mingxuan Du, Tuowen Zhou, Benfeng Xu, Hongtao Xie
Abstract: Text-to-image and personalized editing models now synthesize high-fidelity single-subject images with ease. Yet placing multiple named people into shared contact actions such as embrace, carry, or grapple still exposes major failures: fused limbs, invented extremities, and interpenetrating bodies. Existing evaluations largely overlook these anatomical and geometric issues, and VLM-as-a-judge checklists often saturate on Interaction while the errors remain obvious to humans. We introduce MPIE-Bench, a 2,500-sample benchmark of video-mined editing triplets spanning 405 scenes, 14 interaction categories, and four contact densities (C0-C3). We also propose MPIE-Eval, whose two new axes score contact-time geometry from a frozen public multi-person mesh reconstruction. Anatomy asks whether every human-like mass is explained by a complete set of reconstructed bodies, and Interaction asks whether the penetration and surface distance between those bodies match the contact the instruction asked for. Across ten editors, mesh Anatomy tops out at 0.65 and mesh Interaction at 0.72 on two different models, so no single editor is strong on both, while VLM checklists rate the same images above 0.95. A five-rater study confirms that both axes track human judgement more closely than a zero-shot VLM judge, and the rankings hold under ablation of every weight and threshold.
Authors: Jianwei He, Kailin Lyu, Junhao Dong, Long Xiao, Wenjie Hou, Jingze Lu, Di Wu, Lin Shu, Jie Hao
Abstract: Deep learning has shown strong potential in medical image analysis, but most existing methods rely on large-scale annotations and a closed-world assumption that rarely holds in clinical practice. Although Generalized Category Discovery (GCD) has advanced rapidly on natural images, it remains underexplored in medical imaging. To address this issue, we propose MedXplore, a unified framework for reliable and unbiased medical GCD, optimizing from both perceptual and decision levels. Specifically, at the perceptual level, taking a frequency domain perspective, Frequency-SNR Adaptive Attention and Consistency (FAAC) performs learnable full-spectrum filtering and global-local energy contrast activation to not only highlight local abnormal signals relative to the global context, but also provide reliable semantic anchors for patch consistency learning. At the decision level, Adaptive Cosine-Angular Margin (ACAM) adjusts angular margins using semantic difficulty and feature confidence to balance intra-class compactness and inter-class separability. Together, the two modules improve lesion-sensitive representation learning and mitigate old-class bias. Experiments on multiple benchmarks show an average \textbf{8.5\%} gain in \textit{All} accuracy over the strongest competing methods. On Kvasir, MedXplore reduces false-old errors from 14.50\% to 0.80\%, demonstrating strong robustness under severe old-new ambiguity.
Authors: Mingzhe Lyu, Jinqiang Cui, Hong Zhang
Abstract: Low-light image enhancement (LLIE) methods involve tunable parameters that are typically fixed, often leading to performance degradation when applied across scenes. Manually selecting the best configuration, however, can be time-consuming and not always practical. Peak signal-to-noise ratio (PSNR) is the natural fidelity criterion for automating parameter selection, yet it requires a ground-truth reference that is typically unavailable. To our knowledge, no learning-based method addresses no-reference PSNR prediction for low-light image enhancement; the natural surrogate, no-reference image quality assessment (NR-IQA), targets perceptual quality rather than signal fidelity, and all seven baselines we test achieve 0% top-1 selection accuracy on our benchmark. With paired training data, the ground-truth PSNR is analytically computable, providing exact supervision without a separate teacher network. Building on this, we propose BlindPSNR, a lightweight no-reference network that fuses the enhanced image with the degraded low-light input via windowed cross-attention and estimates PSNR through heteroscedastic regression. While a scalar-regression baseline achieves top-1 accuracy of 54.4%, BlindPSNR raises this to 89.5% with regret dropping from 1.62 dB to 0.026 dB, and generalizes to unseen datasets (SRCC = 0.61-0.67).
Authors: Renlong Wu, Haoran Chen, Yuxiang Wei, Xiaowei Jin, Wangmeng Zuo, Hui Li
Abstract: Generating high-quality 360-degree dynamic human assets from text prompts is challenging. Existing methods usually synthesize monocular or multi-view videos first and then fit a 4D representation, which is expensive and often causes incomplete geometry or view-inconsistent renderings. We present 4DHumanDiff, a diffusion framework that directly generates dynamic humans represented by 4D Gaussian Splatting (4DGS) from text prompts. By modeling the structured 4D representation space end-to-end, 4DHumanDiff avoids video pre-generation and per-scene reconstruction, making it better suited for view-consistent and temporally coherent asset generation. The model uses a 3D U-Net backbone with temporal attention for motion-aware generation. We further construct a large-scale text-to-4DGS dataset with 60,000 high-quality pairs, and introduce 2D regularization and training-free 4D interpolation to improve rendering quality and motion smoothness. Experiments show that 4DHumanDiff generates consistent 360-degree dynamic humans within one minute, achieves better temporal and multi-view consistency, and reduces inference time by more than 10x.
Authors: Wenjie Zhu, Yabin Zhang, Wenjun Zeng, Lei Zhang
Abstract: Multimodal Large Language Models (MLLMs) have achieved strong performance on a wide range of vision-language tasks, but often fail under imperfect or shifted contexts. A reliable MLLM should refuse truly out-of-context (OOC) questions with subject-level context shifts while still answering shifted in-context (Shifted IC) questions with non-subject context shifts. Existing benchmarks mainly target OOC or visually unanswerable questions, but overlook answerable Shifted IC cases and cover limited OOC shifts. To fill this gap, we present MMOOC, a large-scale benchmark for evaluating refusal and robust answering abilities of MLLMs. MMOOC contains over 41K image-question pairs, including answerable Shifted IC cases and unanswerable OOC cases, spanning three question formats, eight shift types and six visual scenarios, with data quality ensured through MLLM-based filtering and human verification. We evaluate model responses using Accuracy and Refusal Rate, and further introduce an LLM-as-a-Judge metric to assess the correctness of model reasoning. Experiments on diverse MLLMs show that current models still struggle to balance answer-ability and refusal under shifted contexts. We further analyze key failure patterns and show that post-training can improve robustness. MMOOC will be made publicly available.
Authors: Chuanzhi Xu, Ziyuan Tao, Jean Julien KNell, Yanrong Chen, Haolan Guo, Xuanhua Yin, Adnan Mahmood, Weidong Cai
Abstract: Personalized image enhancement should reflect individual aesthetic taste, yet learning such preferences commonly depends on private photos and ratings that are unsuitable for centralized collection. The task must infer preference from sparse, heterogeneous feedback and translate it into natural-looking color transformations on resource-constrained user devices. We introduce FedPAIE, a federated personalized aesthetic image enhancement framework for user-adaptive color grading without centralizing raw photos or ratings. FedPAIE trains a lightweight dual-cue aesthetic scorer, calibrates it into a personalized scorer on a small local support set, and freezes it to guide regularized adaptation of a lightweight CLUT enhancer from unpaired local photographs. Fidelity constraints and an excess-gap penalty regularize scorer-guided adaptation to limit proxy-score over-optimization while preserving content and natural appearance. Training remains lightweight throughout the pipeline: scorer learning updates at most 0.787M parameters, enhancer adaptation updates 0.265M, and inference retains only a 0.293M-parameter personalized enhancer. Experiments on MIT-Adobe FiveK and Flickr-AES demonstrate effective open-world personalization and a favorable balance between user preference and image fidelity. FedPAIE thus connects decentralized preference learning with efficient personalized image transformation without requiring paired user retouches.
Authors: Fexiang Liu, Shiye Wang, Qiang Qiu, Zheng Wang
Abstract: Reliable deployment of multimodal large language models (MLLMs) requires deciding whether a confident visual answer should be trusted, reviewed, or routed to a stronger system. Confidence scores capture candidate margins, but not where the estimated signed visual readouts associated with those margins come from or how they are distributed. We study inference-time risk detection for closed visual answers using the same white-box prefill path that produces the answer. Witness Evidence Portfolios (WEP) first estimates, layer by layer, which visual contributions support or contradict the predicted candidate. It summarizes these contributions through two interpretable route families: question-related evidence provenance and signed evidence concentration. Nested grouped validation chooses the more reliable family and a sparse top-k route portfolio, which is fused with candidate confidence. WEP needs no image perturbation, decoding change, backward pass, or external verifier. Across three MLLMs and four binary-answer benchmarks, WEP improves mean error AP by 0.134. All 12 model--dataset gains are positive, and image-cluster bootstrap intervals are strictly positive on 10 pairs. WEP targets white-box closed-answer systems and uses a labeled calibration slice.
Authors: Shawn Li, Wei Yang, Jike Zhong, Jiate Li, Jiawei Yang, You Qin, Ryan Rossi, Franck Dernoncourt, Roger Zimmermann, Yue Wang, Zhengzhong Tu, Vicente Ordonez, Mohit Bansal, Yue Zhao
Abstract: Jigsaw puzzle solving requires jointly reasoning about visual content and geometric constraints, yet existing benchmarks use rectangular cuts that create ambiguous ground truth in texture-repeated regions. We introduce \textit{\ours{}}, a benchmark with tab-and-blank interlocking pieces where geometric constraints provide strong local compatibility requirements that, combined with visual content, yield unambiguous ground truth. Across 95K instances at four grid densities (4$\times$4 to 16$\times$16), we find that \textbf{zero-shot VLMs largely lack geometric reasoning}: only one of five frontier models (GPT-5.5) exceeds random baseline on 4$\times$4 puzzles, while all others perform at chance level. While supervised fine-tuning achieves $>$97\% on 4$\times$4, \textbf{all models collapse on larger grids}: GPT-5.5 drops from 70\% to near-random on 8$\times$8, and even fine-tuned models fall below 5\% on 12$\times$12. This ``scaling cliff'' suggests current architectures cannot maintain consistent constraint satisfaction as the number of pieces increases. \ours{} establishes scalable geometric reasoning as an open challenge for vision-language models.
Authors: Shaobo Liu, Feiqiao Mao, Shuaishuai Zhou, Yan Zhan, Weiqi Tan, Zhiqiong Lu, Zhengping Liang
Abstract: We propose RefineSVG, a single-step closed-loop visual feedback framework that enables multimodal large language models (MLLMs) to perform high-fidelity image-to-SVG generation through self-correction. Existing MLLM-based approaches rely on single-pass open-loop inference, where the model receives visual input only once and must generate thousands of SVG code tokens without intermediate verification. This paradigm inevitably leads to geometric drift, error accumulation, and visual hallucination on complex images. RefineSVG overcomes this limitation by invoking an external rendering engine after an initial SVG generation pass to compare the rendered output against the target image. The comparison yields a multi-dimensional visual residual map (Diff-Map) that is fed back to the model as a ReAct-style correction signal, driving a targeted correction step. To support this render-observe-correct interaction, we further introduce an SVG-oriented semantic vocabulary that compresses token sequences by over 52%. A progressive training pipeline spanning supervised fine-tuning, rejection-sampling cold-start data construction, and end-to-end agentic reinforcement learning aligns the model with closed-loop visual correction. Extensive experiments show that RefineSVG consistently outperforms existing baselines in reconstruction fidelity, structural accuracy, and code efficiency.Code is available at https://github.com/liuxiaobo66/RefineSVG.
Authors: Jiasheng Li, Zhong Ji, Yan Zhang, Huihui Li
Abstract: Large Vision-Language Models (VLMs) suffer from prohibitive inference overhead due to long sequences of visual tokens. However, existing visual token reduction methods mainly improve efficiency by pruning or compressing redundant tokens without examining whether the resulting representation remains semantically consistent with the original representation. Mapping the original N-token visual sequence to K tokens may discard, dilute, or misassign critical visual cues, triggering severe semantic drift that deviates the VLM's understanding. In this paper, we first introduce the principle of 'Calibrate Before Reason' to visual token reduction and propose CaRe, a training-free robust framework that calibrates compact visual representations before reasoning to preserve semantic fidelity in VLMs. CaRe consists of two mutually complementary modules: 1) Perturbation-Robust Calibration Anchoring, which identifies calibration anchors with stable model-side influence under multi-directional perturbations; 2) Confidence-Gated Token Calibration, which extracts reliable calibration signals from unselected tokens and injects them into anchors. Extensive evaluations across diverse VLM architectures and benchmarks verify that CaRe outperforms state-of-the-art token reduction baselines. While pruning 94.4% of visual tokens, our method retains 96.4% of the original full-token performance, delivering up to 2.30 times faster end-to-end inference speed relative to unpruned vanilla models.
Authors: Sangmin Hong, Daniel Sungho Jung, Heewon Kim, Kyoung Mu Lee
Abstract: Point clouds are one of the most fundamental and widely used 3D representations, serving as the most basic geometric representation of 3D shapes. Nevertheless, most existing 3D printing pipelines require a watertight mesh as input, preventing the direct use of point clouds for fabrication. A common workaround is to reconstruct meshes from point clouds; however, the resulting meshes often contain geometric artifacts, such as incorrect faces or topological inconsistencies, that are difficult to repair and may lead to printing failures. To overcome these limitations, we propose PrintAnything, a novel framework that learns to produce executable 3D printing G-code directly from 3D point clouds without requiring mesh reconstruction. To enable point clouds to serve as direct input for slice-wise toolpath generation, we introduce a slice-wise point projection strategy that transforms unstructured 3D point clouds into slice-aligned 2D representations consistent with layer-by-layer nature of fused deposition modeling in 3D printing. To eliminate mesh dependency and provide a unified representation that bridges point clouds and G-code, we propose Geometric plan (G-plan) map, a compact 2D representation composed of occupancy, region, and flow maps that encode the geometric and extrusion properties required for toolpath synthesis in 3D printing. As a result, our proposed method accurately generates printable G-code directly from point clouds, enabling a practical and fully mesh-free pipeline for 3D printing. The code is publicly available at \href{https://github.com/Sangminhong/PrintAnything}{https://github.com/Sangminhong/PrintAnything}.
URLs: https://github.com/Sangminhong/PrintAnything, https://github.com/Sangminhong/PrintAnything
Authors: Daeun Lee, Jaeah Lee, Woosung Kim, Haebeom Jung, Jaesik Park
Abstract: Building interactive digital twins requires recovering both 3D geometry and the kinematic structures that govern how objects articulate. Yet existing methods for articulated object reconstruction require explicitly observable motion from multiple articulation states. We introduce a rest-state formulation that reconstructs articulated objects from a single closed configuration, an inherently ill-posed setting where geometry, semantics, and motion priors compensate for the absence of motion cues. Our framework adopts an explicit mesh as an intermediate representation for cross-model verification and fusion, reconciling noisy outputs from vision-language and segmentation models into spatially consistent part structures. To estimate joint parameters without observed motion, we use a video diffusion model to synthesize articulation hypotheses and validate them through geometric consistency. Our approach achieves accurate part decomposition and physically plausible articulation, performing competitively with motion-observing reconstruction-based, generation-based, and modular pretrained-model baselines.
Authors: Jaehun Jung, Wonjun Kim
Abstract: We address the problem of recovering the full-body mesh from only the head pose. This task has become essential for various applications based on head-mounted devices or smart glasses. The challenge of this task lies in estimating the pose information of unobserved body parts based solely on a single joint (i.e., head) trajectory. Several studies have begun to adopt head-conditioned generative models, however, such previous methods are costly and time-consuming due to the diffusion-based iterative process. As an alternative, we propose a simple yet novel method that leverages the latent space of the guidance network, which is designed as a variational autoencoder taking full-body poses as inputs. By enforcing latent distributions of this guidance network and our head-to-motion network to be similar, latent features sampled from the 'guided' distribution, i.e., distribution learned in our head-to-motion network, can be reliably decoded for natural representations of full-body poses even only with the head pose. One important advantage of the proposed method is that one-step sampling scheme achieves remarkably fast inference (more than 50 times faster) compared to diffusion-based approaches. Experimental results on benchmark datasets show that the proposed method efficiently improves the performance of ego-body mesh reconstruction.
Authors: Bowen Wang, Youwen Zhang, Ritesh Mehta
Abstract: We describe the DS@GT submissions to the ImageCLEFmedical Caption 2026 challenge, which continues a long-running benchmark on the ROCOv2 dataset with two tracks: Concept Detection (Task 1), assigning UMLS Concept Unique Identifiers (CUIs) to radiology images, and Caption Prediction (Task 2), generating natural-language captions. For Task 1, our primary submission was a three-way late-fusion ensemble of ConvNeXt-V2, BiomedCLIP ViT-B/16, and DenseNet-169 with a regularized ''Honest Threshold Tuning'' procedure designed to avoid validation overfitting on rare concepts; this submission ranked first on the official submission with a primary $F_1$ of $0.5790$ and a secondary $F_1$ of $0.9657$. In parallel, we submitted a training-free KNN retrieval pipeline over frozen BiomedCLIP embeddings, which reached a primary $F_1$ of $0.5780$ and a secondary $F_1$ of $0.9599$-essentially matching the fine-tuned ensemble on the primary track at a fraction of the cost. For Task 2, our submissions included a fine-tuned Gemma-3 27B model (overall $0.3571$, ranking third in the official submission), a fully fine-tuned BLIP pipeline with custom Vizwins merging ($0.3564$), and a zero-shot MedGemma-4B run with a PubMed-style prompt ($0.3186$), spanning a wide range of model scales and training costs. Code: https://github.com/dsgt-arc/imageclef-caption-2026.
Authors: Shuhuan Chen, Xiangyu Zhu, Weisong Zhao, Siran Peng, Tianshuo Zhang, Haoyuan Zhang, Haichao Shi, Xiao-Yu Zhang, Zhen Lei
Abstract: Publishing private face recognition~(FR) training datasets is privacy-sensitive because faces expose identity information. Private FR training dataset publication mitigates this risk by releasing protected proxies as substitutes for private training faces. However, training FR models with such data introduces an identity paradox: \emph{the identity cues that make released faces useful for recognition supervision are also the cues that make them linkable to real individuals.} A protected face should be decoupled from the original identity, yet still behave as a reliable identity sample for training. Removing these cues too aggressively may destroy the class structure needed for recognition learning, whereas preserving them too faithfully may increase source-identity linkability. We argue that this paradox stems from conflating source-aligned identity semantics with recognition-useful proxy identity geometry. The former should be suppressed to reduce linkage to private individuals, while the latter should be preserved for FR learning. Based on this insight, we propose \textbf{Private Face Distillation}, an identity-decoupling and geometry-preserving framework. It uses Orthogonal Geometry Preservation to construct decoupled proxy identities from private identity representations while maintaining hyperspherical geometry, and Relational Topology Alignment to preserve identity relations for recognition learning. Experiments across multiple domain-shifted FR scenarios show that Private Face Distillation achieves stronger utility than the evaluated publication baselines. On IJB-C surveillance, it improves $\mathrm{TAR}@\mathrm{FAR}{=}1\text{e-}{3}$ by 3.94\% over the baseline while reducing source-identity linkability. These results suggest that private FR training dataset publication should decouple source-identity correspondence while preserving proxy identity geometry.
Authors: Tomer Erez, Moshe Kimhi, Chaim Baskin, Ehud Rivlin
Abstract: Large chest radiography archives are difficult to search because most studies are paired only with free-text reports rather than structured clinical annotations. Vision-language models offer a natural interface for text-to-image retrieval, but current biomedical models are primarily optimized for report-to-image matching rather than for satisfying short clinical search queries. This creates an objective mismatch: a model may retrieve images related to words in the query while failing to satisfy the full clinical constraint, especially for conjunctions and negations such as ``atelectasis and no pneumonia.'' We introduce CXR-Retrieve, a structured benchmark for compositional chest X-ray text-to-image retrieval. The benchmark contains 5,159 test images from the official test-split of MIMIC-CXR-JPG and 145 textual queries spanning single and conjunction findings, both positive and negative. Relevance is defined by whether a retrieved image satisfies all asserted pathology constraints, rather than by whether it matches a paired report. We further propose a label-aware contrastive fine-tuning objective for clinical retrieval. Our method attracts image-text pairs with compatible asserted pathology constraints, including shared confirmed absences, while explicitly repelling contradictory pairs. Starting from the in-domain CXR-CLIP checkpoint, our method improves Precision@5 over CXR-CLIP by 8.5 percentage points on two-pathology conjunctions and by 22.0 percentage points on negation queries. These results show that reliable chest X-ray retrieval requires training objectives that model not only which findings are mentioned, but also how they are clinically asserted.
Authors: Chunpeng Wang, Yuxin Li, Xiaoyu Wang, Jidong Yang, Suo Gao, Qi Li
Abstract: Existing invisible watermark removal methods often struggle to accurately capture the watermark-bearing features, leading to an unfavorable trade-off between watermark suppression and perceptual fidelity. In this paper, we propose the Frequency-Decoupled Diffusion Watermark Attack Network (FDDWAN), a coarse-to-fine framework that performs watermark removal through wavelet-domain decomposition and residual diffusion refinement. In the initial stage, the Wavelet-based Frequency-domain Preliminary Attack Module (WFPAM) decomposes the watermarked image into low- and high-frequency subbands and applies frequency-specific attack strategies tailored to their respective contributions to watermark robustness and perceptual quality. In the next stage, the Frequency-domain Residual Diffusion Attack Module (FRDAM) separately models the residual distributions between the preliminarily attacked outputs and the corresponding watermark-free references during training. Rather than reconstructing the entire image, FRDAM selectively refines frequency-domain residuals, directing the diffusion process toward the remaining watermark related discrepancies while minimizing modifications to image content. Extensive experiments on CelebA and ImageNet across four representative watermarking schemes demonstrate that FDDWAN achieves a more favorable trade-off between watermark removal effectiveness and visual fidelity than conventional and learning-based attack methods.
Authors: Zhilin Wu, Zhangkai Ni, Chengmei Yang, Longzhen Yang, Yihang Liu, Ying Wen, Lianghua He
Abstract: In clinical practice, patients often undergo multiple imaging examinations over successive visits, yielding longitudinal data. Modeling such temporal information is crucial for reliable assessment of disease progression and treatment response. However, despite the rapid advancement of multimodal large language models (MLLMs), longitudinal medical visual reasoning remains largely underexplored. To fill this gap, we propose LoMeVQA, a comprehensive benchmark consisting of 206K longitudinal visual question answering (VQA) pairs for temporal medical image analysis. LoMeVQA covers five tasks: progress classification, progress description, progress report generation, differential region grounding, and differential region description. To construct the dataset, we develop an automated pipeline that (1) organizes patient records chronologically, (2) extracts clinically meaningful entities via a medical knowledge graph, and (3) models their temporal evolution to guide large language models in generating high-quality longitudinal VQA pairs. Extensive evaluations demonstrate that both general-purpose and medical-domain MLLMs perform poorly on LoMeVQA, revealing substantial limitations in temporal reasoning. To address these limitations, we introduce MedLong-8B, which achieves state-of-the-art performance across all tasks. Beyond benchmarking, we conduct detailed analyses that uncover key failure modes and shed light on how to improve longitudinal medical visual reasoning. Our data is available at: https://github.com/pepperbubble/LoMeVQA
Authors: Chunpeng Wang, Yanan Shi, Zhiqiu Xia, Jidong Yang, Suo Gao, Qi Li
Abstract: Existing watermark attacks typically rely on predefined signal-processing operations or locally constrained restoration networks, making it difficult to capture the long-range dependencies of globally distributed watermark signals and resulting in an unfavorable trade-off between removal effectiveness and visual fidelity. In this paper, we propose SPFM-Net, a semantic-prior-guided and frequency-constrained Mamba framework for invisible watermark attack. SPFM-Net first employs high-ratio masking to disrupt the spatial coherence of invisible watermark signals, and then utilizes a partially fine-tuned pretrained Masked Autoencoder to reconstruct semantically consistent image from sparse observations while suppressing watermark-related information. A Multi-scale Residual Frequency Feature Interaction module subsequently aggregates watermark-related residual features across multiple receptive fields, while adaptively suppressing responses from watermark-irrelevant regions. To further capture the long-range dependencies of globally distributed watermark signals, a lightweight Mamba-based Global State-space Feature Modeling (GSFM) unit is introduced to separate watermark-related features from natural image content and suppress the remaining watermark traces. In addition, SPFM-Net is optimized using a multi-level objective that jointly imposes spatial-, frequency-, and edge-domain constraints, enabling effective watermark suppression while preserving perceptual quality. Extensive experiments on representative spatial-domain, transform-domain, orthogonal moment-based, and deep learning-based watermarking schemes demonstrate that SPFM-Net achieves a favorable trade-off between watermark attack effectiveness and perceptual fidelity.
Authors: Lei Yang, Xinze Liu, Dayan Wu, Ding Wang, Hengjie Zhu, Zihao Zhang, Tianzhu Hu, Hanqi Wu, Peng Fu, Zheng Lin
Abstract: Large vision-language models (LVLMs) often hallucinate objects that are absent from an image. Despite recent progress, existing mitigation methods still lack reliable object-level grounding diagnostics and therefore tend to apply coarse-grained interventions, which can impair visual understanding, shorten responses, and reduce coverage of genuinely grounded objects. The key challenge is thus to detect, during generation, whether each emerging object mention is supported by reliable visual evidence, so that hallucination can be mitigated selectively. Yet output confidence reflects next-token plausibility rather than visual support, allowing language priors to make absent objects appear certain. We show that the missing diagnostic evidence is encoded in an Intrinsic Grounding Signature (IGS), a distributed signed attention pattern that remains informative for such confident hallucinations. Based on IGS, we propose Verifier-Guided Decoding (VGD), a decoding framework in which a lightweight verifier examines each emerging object mention, rolls back the KV cache when the mention is identified as high risk, suppresses the object and its synonyms, and regenerates the affected continuation. Because VGD intervenes only on object mentions identified as high risk, it reduces object hallucination while preserving the model's original visual understanding and grounded object coverage. Experiments on CHAIR and AMBER-G show that VGD achieves state-of-the-art object hallucination reduction: at @rec90, it cuts AMBER-G CHAIR by 43.6\% while retaining 99.6\% of grounded-object coverage, and reduces CHAIR-MSCOCO CHAIR$_i$/CHAIR$_s$ by 37.0\%/30.4\% without shortening captions.
Authors: Zhongkuan Mao, Xianjie Liu, Tianyu Meng, Yidong Wang, Wenzhuo Zhao, Ronghao Xian, Yao Jiang, Fei Shen, Junfeng Fang, Yong Dai, Yi Zhang, Keren Fu
Abstract: High-resolution visual question answering (HR-VQA) is often treated as a problem of insufficient evidence acquisition, where failing multimodal large language models must inspect images again through cropping, re-encoding, or multi-round search. We show that this view is incomplete: in many cases, fine-grained evidence has already survived visual encoding and become identifiable and influential within an intermediate-layer routing window, but is later diluted before answer generation. We propose Thinking-Once, a \textbf{training-free, single-visual-pass} evidence-routing method that reconstructs question-conditioned attention at this window, preserves core entity tokens and compact background context, and routes this evidence to later layers without extra visual encoding. Across five base models, Thinking-Once consistently improves or matches the corresponding base setting, increasing the average scores on V$^*$Bench, HRBench-4K, and HRBench-8K by \textit{+3.1}, \textit{+3.0}, and \textit{+2.7} points while reducing the average peak memory by about 4,GB. On Qwen2.5-VL-7B, it improves the three benchmarks by \textit{+9.9}, \textit{+4.6}, and \textit{+5.5} points, raising the cross-benchmark mean from 72.5 to 79.1. With the ZwZ-8B base model, Thinking-Once reaches a mean score of 82.7. Against 11 open-source HR-VQA baselines, it obtains the best or tied-best score on all three benchmark averages and the best overall mean; for example, compared with DeepScan, it reduces V$^*$Bench inference time by \textbf{97.2\%} while improving the cross-benchmark mean from 77.8 to 79.1. These results show that HR-VQA can be improved by routing already encoded evidence rather than repeatedly acquiring new visual inputs. Code is available in the appendix.
Authors: Harshit Mittal, Arash Rabbani
Abstract: Accurate cell segmentation and classification are foundational to digital pathology, enabling quantitative tissue analysis for diagnosis and treatment planning. Encoder-decoder architectures that fuse multi-scale features through skip connections have become the dominant paradigm for this task, yet standard direct skip connections treat every spatial location equally, which leads to redundant and potentially conflicting information reaching the decoder. To overcome this problem, various gating mechanisms have been introduced, but most of them operate solely on filtering encoder information, neglecting the benefit of global contextual information from the decoder. This study proposes replacing conventional skip connections in a CellViT-based model with a novel Spatial Attention Fusion (SAF) Gating module. Each SAF gate concatenates the encoder skip and upsampled decoder features, compresses them through two pointwise convolutions with an intermediate ReLU, and applies a channel-wise softmax to produce a per-pixel "heatmap of trust" that sums to unity at every spatial location, allowing the network to learn where each source is most trustworthy. The resulting fused features improve the model's ability to detect the minority "Dead" class, which in turn enhances the multi-class panoptic quality (mPQ) on the PanNuke dataset. SAF Gating is compared against six gating alternatives including no gating, attention gates, squeeze-and-excitation, CBAM, cross-attention, and attentional feature fusion on PanNuke and MoNuSeg datasets. SAF Gating achieves the highest mPQ (0.471), a gain driven primarily by a 14.5-point improvement in Dead-class F1 score compared to ungated CellViT baseline.
Authors: Hanshuai Cui, Zhiqing Tang, Zhi Yao, Qianli Ma, Fanshuai Meng, Weijia Jia
Abstract: Diffusion models are widely used to generate high-quality images and videos, but their iterative denoising process remains computationally intensive. A growing class of training-free accelerators reduces this cost by reusing cached intermediate features or forecasting future ones. To control draft drift, these methods sometimes compute an exact block feature for verification. Yet the resulting exact feature is typically used only to measure discrepancy or guide a later decision and is then discarded. We find that this previously computed feature can instead be reused for correction. Forwarding it at the verification site resets the local draft residual and reduces downstream feature error. Based on this observation, we introduce FeatFix, a local exact-feature correction method for cached diffusion inference. FeatFix operates at a fixed sparse set of layer--timestep sites. At each selected site, it replaces the complete draft block output with the exact output computed from the same incoming state, avoiding token- or channel-level partial replacement and full-timestep recomputation. Experiments across four image and video backbones show that FeatFix consistently accelerates generation, achieving a speedup of up to $6.70\times$ over Vanilla while maintaining competitive output quality.
Authors: Songsong Duan, Xi Yang, Nannan Wang
Abstract: Camouflaged Object Detection (COD) aims to identify and segment camouflaged objects in complex environments, which are often concealed because their color and texture are similar to the background. Several existing COD methods introduce depth maps to boost detection performance via learning complementary RGB-D features, ignoring modality-specific characteristics of concealed objects in the depth domain. To address this issue, we propose a depth collaborative network, called VCP-DCN, to mine distinguishable multi-modality features beyond visual concealed prototype in depth domain. Specifically, VCP-DCN progressively performs multi-modality alignment, interaction, and fusion for the COD task. In the \textbf{alignment} stage, we propose a Separable Prototype Embedding (SPE) module to learn modality-consistency and modality-specific RGB/depth prototype tokens through prototype contrastive learning. Furthermore, we develop a Multi-modality Dual Attention (MDA) module to enhance the cross-modal feature representation through local response maps between modality-consistency RGB/depth prototype tokens and visual tokens on the \textbf{interaction} stage. Finally, we design a Depth Adaptive Injection (DAI) module to adaptively measure contribution of RGB/depth features with a decision-making mechanism, which calculates similarity distance between RGB/depth modality-specific prototype tokens and modality-consistency ones on the \textbf{fusion} stage. Extensive experiments demonstrate the effectiveness of our VCP-DCN on three authoritative datasets.
Authors: Jinghong Liu, Yuchuan Deng, Fanping Liu, Meng Huang, Xirong Li
Abstract: Few-shot medical image segmentation (FS-MIS) aims to segment novel regions of interest (ROIs) from a few annotated support examples. Despite rapid progress, existing FS-MIS solutions span diverse paradigms but are evaluated under inconsistent settings, leaving their relative effectiveness unclear. We introduce FAME, a unified benchmark for evaluating FS-MIS solutions, covering specialists, SAM-based methods, CLIP-based methods, and MLLM-based methods. FAME contains 14,958 test samples across 7 anatomical sites, 9 imaging modalities, and 14 ROI categories, and evaluates models under zero-shot and ten-shot settings with additional assessment of target-absence recognition and generalization under covariate and semantic shifts. Our evaluation reveals several findings. First, effective few-shot segmentation depends on how models exploit support examples: direct visual adaptation generally outperforms prompt-based strategies. Second, increasing support examples improves performance only when models can effectively utilize them. Third, semantic transfer remains substantially more challenging than imaging-domain adaptation, and strong localization ability does not necessarily imply reliable target-absence recognition. We hope FAME provides a comprehensive understanding of current FS-MIS solutions and facilitates the development of more effective and reliable few-shot medical segmentation methods.
Authors: Kaifan Zhang, Lihuo He, Yuqi Ji, Junjie Ke, Lukun Wu, Tianhao You, Xinbo Gao
Abstract: Recent EEG-to-image retrieval models have achieved strong performance in identifying viewed images from semantically diverse candidates. Yet such success does not reveal what visual information supports the match. A model may readily identify a cheetah among tools, plants, and vehicles, but can it still distinguish the viewed cheetah from the same scene with the cheetah replaced by a dog? Motivated by this question, we introduce EEG-EditBench, a diagnostic benchmark that examines this question through controlled edits of object identity, attributes, background, and object presence. Built from the 200 THINGS-EEG2 test images, EEG-EditBench contains 2,137 quality-controlled edits and evaluates eight representative EEG visual decoding models. Our results show that strong standard retrieval does not consistently transfer to edit-based evaluation, with fine-grained attribute changes presenting the greatest challenge. EEG-EditBench reveals model behavior hidden by aggregate retrieval accuracy and provides a controlled basis for studying what visual information EEG-image models preserve. The code and complete dataset are publicly available.
Authors: Dragos Costea, Alina Marcu, Cristina Lazar, Marius Leordeanu
Abstract: Perceiving human motion and intent at long range is a prerequisite for socially intelligent aerial robots, yet the data to learn it barely exists. We introduce Drones2BodyLanguage, a dataset grounding human motion in real UAV footage: avatars manifesting ten communicative intents are placed into unmodified 4K drone scenes with metrically correct position, scale and orientation, maintained over hundreds of frames of camera motion. Enabling it is a lightweight geometric world model of the local scene - semantically selected anchors lifted to 3D through streaming monocular depth - in which a placement point is predicted as an affine anchor combination with provably rigid-invariant weights, and re-rendered under an SVD-fitted ground rotation. Across twelve architectures on scene- and motion-disjoint splits, training on placed data lifts mean intent accuracy by a wide margin for real, retargeted and generated motion alike, with gains confirmed on two in-the-wild scenes.
Authors: Zihao Cai, Xinghan Li, Ruiyan Yang, Xue Song, Haijun Shan, Jingjing Chen
Abstract: As generative models continue to evolve, AI-generated image detectors must incrementally adapt to emerging generative domains while preserving knowledge acquired from previous ones. This continual learning setting is particularly challenging because forensic traces are often subtle and generator-specific, making detectors highly vulnerable to catastrophic forgetting. Existing methods primarily address this problem by stabilizing feature representations, implicitly treating forgetting as a representation-level issue. In this paper, we show that this perspective is incomplete. We demonstrate that even when feature representations remain discriminative, the decision boundary can progressively drift as the classification head is continually optimized on new domains. These two effects jointly give rise to a compound failure mode, termed Dual Degradation. To overcome this challenge, we propose DECODE, a decoupled continual detection framework that jointly mitigates representation- and decision-level forgetting. Specifically, we introduce Subspace Diversity Regularization (SDR) to preserve diverse forensic representations and Closed-Form Decision Alignment (CDA) to recalibrate the shared classification head after each adapter merge without manual hyperparameter tuning. Extensive experiments on 19 generative domains show that DECODE achieves an average accuracy of 99.36% with only 0.39% forgetting, while further generalizing to 11 unseen generators with 95.36% accuracy.
Authors: Nguyen Duc Thai, Junhao Dong, Sua Qi Rong, Hua Yu, Yew-Soon Ong
Abstract: Vision-language models (VLMs), such as CLIP, are vulnerable to adversarial attacks, posing a serious problem for real-life applications and deployment. Adversarial fine-tuning emerges as a prominent defense method; however, different fine-tuning strategies often produce specialized models with distinct robustness characteristics. Each fine-tuned model in turn thrives in some evaluation settings but falters on others, limiting their defensive capabilities. We refer to these specialized fine-tuned models as robust model experts and propose a collaborative adversarial fine-tuning framework: CARE - Collaborative Adversarial Robustness fine-tuning using Embedding alignment. CARE maintains multiple experts during training, enables knowledge exchange through embedding-space harmonization, and consolidates the learned knowledge into a single unified robust model. Experts benefit from one another while preserving their individual specializations, enabling the final model to inherit complementary robustness properties. In this paper, we demonstrate CARE on two different adversarial fine-tuning strategies with complementary robustness behaviors. Extensive experiments on classic image classification and downstream vision-language tasks display the effectiveness of our approach, with CARE being able to outperform individually learned model experts. The results suggest that collaborative learning across model experts is a promising direction for improving adversarial robustness.
Authors: Zaiyan Zhang, Qiangqiang Yuan, Jie Li, Ziyang Lihe, Yu Wan, Yuzeng Chen, Xin Su, Liangpei Zhang
Abstract: Remote sensing images acquired by unmanned aerial vehicles (UAVs) and satellites are often degraded by adverse weather, illumination variation, and imaging artifacts, which may co-occur and jointly induce global distribution shifts and local structural corruption. Although All-in-One image restoration offers an appealing unified alternative to task-specific pipelines, existing methods still suffer from weak or implicit degradation cues and parameter redundancy caused by full-rank multi-expert designs with overlapping restoration behaviors. We propose CoRE-UIR (Common and Residual Experts for Universal Image Restoration), a prior-guided global-local framework centered on the Common-and-Residual Expert Block (CoRE). CoRE explicitly decomposes restoration capacity into a common dense expert for degradation-invariant restoration and low-rank residual experts for degradation-specific compensation, enabling adaptive specialization without redundant expert replication. Built on this design, Degradation Prior Embedding (DPE) adapts frozen CLIP features into an explicit restoration-oriented prior, while Global Feature Modulation (GFM) aligns global feature statistics before local residual compensation. We also construct MDVD-108K (Multi-Degradation VisDrone), a large-scale UAV restoration dataset covering both single and compound degradations, together with a real-world test set. Extensive experiments on multiple datasets show that CoRE-UIR improves the overall average PSNR by 1.05 dB while running 11.83$\times$ faster and reducing peak memory by 85.3% relative to the strongest baseline, BaryIR, thereby maintaining a favorable quality-efficiency trade-off. Evaluations on downstream tasks and unseen degradation also validate the generalizability of CoRE-UIR. The code and dataset will be released at https://github.com/zzaiyan/CoRE-UIR.
Authors: Rui Tang, Wentao Yang, Peirong Zhang, Yongxin Shi, Shun Zhang, Huiguo He, Lianwen Jin
Abstract: Scene text spotting requires high-precision alignment between textual recognition and spatial localization. While visual-token grounding has emerged as a promising formulation for Multimodal Large Language Models (MLLMs), the previous multi-patch paradigm often introduces redundant noise and localization ambiguity, particularly for dense or small text instances. To address this, we propose Single-Patch Text Spotting (SPaTS), a vision-centric framework that routes each text instance through a single anchor visual token and then recovers geometry via full-image refinement. To accurately identify this anchor without oracle labels, we introduce Single-Patch Selective Optimization (SPaSO), a reinforcement learning framework that optimizes discrete visual-token selection using patch-level rewards. To further improve representation robustness and localization precision, we introduce Directional Embedding Alignment (DEA) to suppress unstable norm bias by decoupling feature magnitude and direction, and Patch-Enhanced Decoding (PED) to fuse the routed anchor with language semantics and cross-attend over the full-image feature map for geometry-aware boundary regression beyond coordinate-space surrogates. Extensive experiments demonstrate that SPaTS consistently and significantly outperforms both frontier closed-source MLLMs and OCR MLLMs. Code will be released soon.
Authors: Dongfu Yin, Rourou Su, Cong Zhao, Fei Yu
Abstract: Reflection symmetry detection remains challenging due to interference from asymmetric regions and arbitrary orientations of symmetric patterns. Asymmetric regions introduce background clutter that disrupts symmetric pattern matching, whereas conventional convolutional neural networks lack rotation equivariance, leading to inconsistent feature representations under rotational transformations. To address these issues, we propose an Asymmetric Region Denoising (ARD) module and a Rotation Equivariant Feature Similarity Matching (REFSM) module. The ARD module suppresses asymmetric interference to refine symmetric patterns, while the REFSM module enhances rotation equivariance through feature similarity matching between original and rotated images. Specifically, our dual-input REFSM framework leverages rotation loss to maximize consistency between the score maps of original and rotated images, thereby enabling precise prediction of rotation-equivariant symmetry axes. Furthermore, we introduce GMSYM, a new benchmark dataset that categorizes images into diverse scenarios and incorporates various interferences to address the limitations of existing reflection symmetry detection benchmarks. Extensive experiments on four standard datasets (DENDI, NYU, LDRS, SDRW) and our proposed GMSYM dataset demonstrate that our method achieves state-of-the-art performance in both accuracy and robustness.
Authors: Feng Yang, Xinrui Ju, Keyang Zhang, Xiandong Meng, Rongqun Lin, Howard Leung, Shiqi Wang, Haoliang Li, Chris Xing Tian
Abstract: Multimodal foundation models are reshaping edge-cloud visual intelligence from task-specific feature pipelines into token-based interfaces, where edge devices encode visual inputs into tokens for a general-purpose cloud MLLM. However, dense visual-token sequences increase cloud-side inference costs. Existing pruning methods mainly target centralized inference: vision-driven methods can operate before cloud execution but are typically query-agnostic, whereas query-guided methods often rely on internal states of the target MLLM and cannot determine token relevance before transmission. Compact guidance models offer an alternative, but existing designs may require costly attention aggregation or auxiliary generation. We propose LAST, a training-free framework for query-dependent visual token pruning in edge-cloud collaborative MLLM inference. LAST uses a compact edge-side VLM as a guidance proxy and derives a lightweight importance signal from the last query token's attention to visual tokens. Under causal attention, the last query token can attend to the full visual sequence and the entire query context, enabling query-aware pruning without cloud-model access, autoregressive generation, or costly aggregation over multiple query positions. LAST then retains a diverse set of query-relevant visual tokens under a fixed token budget. We evaluate LAST on 11 multimodal benchmarks under multiple token budgets against pruning methods with different guidance strategies. Experiments show that LAST consistently achieves the strongest performance, preserving 95.4% of the full-token accuracy while retaining only 12.5% of the visual tokens, with low edge-side selection overhead and reduced cloud-side computation.
Authors: Bohan Hou, Haoqiang Lin, Xuemeng Song, Haokun Wen, Meng Liu, Yupeng Hu, Xiangyu Zhao
Abstract: Due to their strong generalizable multimodal processing and reasoning capabilities, Multimodal Large Language Models (MLLMs) have demonstrated significant potential as universal image retrievers, effectively addressing diverse real-world image retrieval tasks. Nevertheless, pioneering studies, while promising, overlook the potential of fine-grained context modeling and disentangled fine-tuning objectives in enhancing MLLMs' retrieval performance, particularly for complex tasks such as long-text-to-image retrieval, visual dialog retrieval, and composed image retrieval (CIR). Therefore, in this work, we propose an automated fine-grained multimodal quintuple dataset construction pipeline and a novel two-stage fine-grained multimodal fine-tuning strategy. The dataset generation pipeline produces a comprehensive CIR dataset with fine-grained image captions and modification text, facilitating fine-grained context modeling. Beyond the previously entangled fine-tuning paradigm, our approach separates the fine-tuning process into two distinct stages: (1) fine-grained context reasoning-oriented fine-tuning and (2) fine-grained retrieval-oriented fine-tuning. These stages aim to sequentially enhance the model's context understanding and query-target alignment capabilities, thereby improving retrieval performance. Extensive experiments across five datasets encompassing diverse and complex image retrieval tasks demonstrate the remarkable superiority of our method over existing approaches in zero-shot retrieval settings, even with a more lightweight MLLM backbone compared to those methods.
Authors: Haotian Zhang, Hao Chen, Han Guo, Zhengxia Zou, Zhenwei Shi
Abstract: Despite substantial progress in remote sensing multi-temporal change detection (MTCD), most existing MTCD methods still represent the dynamic process at each spatial location over the entire observation period using a single change category associated with the final observation. This implicit single-change assumption limits their ability to characterize regions of recurrent change closely related to human activities. To address this limitation, we introduce Urban Building Dynamics Detection (UBDD), which identifies building-change dynamic footprints, i.e., the temporal intervals in which changes occur, from multi-temporal imagery and produces pixel-wise classification masks. For regions undergoing two or more changes, UBDD introduces an independent multi-change class for unified representation, thereby enabling unified modeling of single- and multi-change processes. Furthermore, we propose FootprintNet, which abstracts building-change processes as interactions between latent states and actions, and imposes state-action transition constraints to guide the learning of causally coherent change trajectories. It further exploits temporal change-boundary cues to enhance feature contrast across boundary sides, thereby improving the discrimination among different dynamic footprints and enabling accurate detection of dynamic footprints. Moreover, we introduce the Building Change Dynamics Score (BCDS) to address the inability of conventional metrics to reflect the temporal proximity between predicted footprints and labels. It evaluates predictions according to their preservation of change semantics and temporal offsets from the corresponding labels. Extensive experiments on TSCD, MUDS, and WUSU demonstrate that FootprintNet outperforms current state-of-the-art methods. The code is available at https://github.com/zmoka-zht/FootprintNet.
Authors: Danilo Danese, Angela Lombardi, Tommaso Di Noia
Abstract: The ASNR-MICCAI BraTS Local Synthesis (Inpainting) task asks for the anatomically plausible completion of healthy brain tissue within a masked region of a T1-weighted MRI, providing a tumor-free anatomical reference for downstream analysis. As the task is scored by distortion metrics (SSIM, PSNR, MSE), we build a deterministic regression model and focus on giving it inductive biases tailored to inpainting. Our network follows the U-DiT principle of performing self-attention on a downsampled token grid: a volumetric encoder-decoder imports long-range context through a downsampled global self-attention block with three-dimensional rotary position embeddings, while convolutions and skip connections preserve high-frequency detail. Two ideas drive our results. First, we constrain the attention so that occluded ("void") tokens attend only to known-healthy tokens of the same volume, with a learned bias toward each query's contralateral homologue, forcing the completion to be inferred from observed anatomy rather than from other unknown regions. Second, we add a contralateral-symmetry input that supplies the mirrored healthy hemisphere as a patient-specific prior; since the brain is approximately bilaterally symmetric and lesions are typically unilateral, this prior improves the distortion metrics at matched structural similarity. On the official BraTS-2026 validation leaderboard our submission reaches a mean healthy-region SSIM of $0.864$, PSNR of $24.7$\,dB and MSE of $4.6{\times}10^{-3}$ over $219$ cases. We further analyse the residual smoothness inherent to distortion-optimal regression and discuss its implications for anatomical realism.
Authors: Zihan Qin, Mingze Sun, Yifan Mao, Jialei Xu, Jingfeng Guo, Changrong Hu, Wenbo Zhao, Junjun Jiang, Xianming Liu
Abstract: Rigging is inherently task-dependent because the same mesh may require different skeletons and deformation behaviors across animation tasks. In practice, artists often inspect an initial rig and repeatedly edit its skeletal structure and deformation behavior to meet specific animation requirements. Existing automatic methods primarily generate a plausible rig from geometry, offering limited explicit control over the resulting skeleton and deformation behavior. In this work, we present ViP-Rig, a visual-prompted framework that supports both prompt-first rigging and result-guided editing by injecting features extracted from user-drawn or edited 2D skeletal and rigidity prompts into frozen pretrained backbones. Specifically, ViP-Rig consists of two stages, Skeleton Generation and Skinning Prediction. In the first stage, the skeletal sketch is processed by the Dense-to-Compact Visual Prompt Encoding to produce compact, fixed-length conditioning tokens. The resulting tokens are injected into a frozen pretrained autoregressive generator through gated adapters to control joint placement and branching structure while preserving the generator's geometric prior. In the second stage, the rigidity map is processed using the same visual encoding design, while the pretrained skinning backbone remains frozen. The resulting tokens are symmetrically injected into the point and joint streams to modulate point-joint compatibility and the resulting skinning weights. Experiments on Articulation-XL2.0 and zero-shot evaluation on ModelsResource show that ViP-Rig more accurately recovers target skeletons and skinning weights than geometry-conditioned baselines under prompt-guided evaluation. Qualitative results further demonstrate explicit and localized control in both prompt-first rigging and result-guided editing.
Authors: Zaki Mahfoud, Juan A. Chiavassa, Simon Walther, Florian Haselbeck, Ehsan Yaghoubi
Abstract: Declining insect populations make reliable biodiversity monitoring increasingly urgent, yet monitoring of insect biodiversity is hampered by a lack of standardised data and by costly and time-consuming manual identification by expert entomologists. Deep learning-based image classifiers, processing data from automated non-lethal camera traps, have the potential to transform and scale insect biodiversity monitoring. However, challenges remain in acquiring expert-annotated datasets, developing model architectures that generalise well across diverse taxonomic levels and training models on highly imbalanced data. Hierarchical data also benefits from designing models that default to higher-confidence, coarser-level predictions, when uncertain about finer taxonomic levels. In this paper we address these challenges with a deep learning-based hierarchical classification model. First, we present a manually curated, long-tailed dataset of around one million images of insects, extracted from 1,801 camera-trap video recordings and annotated with a five-level, 34-class hierarchy. Further, we adapt a hierarchical classification model architecture to a five-level variable-depth hierarchy, with class-balanced weighting. Our model improves on non-hierarchical classifiers by leveraging biological taxonomy to extract granularity-specific visual features and makes hierarchy-consistent predictions to the deepest taxonomic level that meets a confidence threshold (T = 0.6). Our model achieved a per-level accuracy of 80-99% on test data, across five levels of hierarchy. Furthermore ...
Authors: Sweta Banerjee, Alireza Teimoury, Nils Porsche, Alexandra K. Stoll, Viktoria Weiss, Niklas Hargarter, Jonas Ammeling, Thomas Conrad, Christoph Stroblberger, Christopher Kaltnecker, Robert Klopfleisch, Christof A. Bertram, Katharina Breininger, Marc Aubreville
Abstract: Pathology foundation models (FMs) are models trained on vast amounts of typically unlabeled data and have been shown to yield regularized latent spaces that can be used effectively in downstream classification tasks. This is also true for the classification of mitotic figures vs. other cells. However, it is so far unclear if the latent space of current FMs provides features that are discriminant and spatially suitably resolved to also serve as a backbone for dense object detection paradigms. In this work, we investigate this question for common current pathology FMs (UNI, UNI2-h, Virchow, Virchow2, H-optimus-0, H-optimus-1) and compare their performance against a fully end-to-end trained baseline based on a ResNet50 architecture. We combine FM backbones with representatives of single stage, dual stage and self-attention-based detectors (RetinaNet, Faster R-CNN, Deformable DETR respectively) on the multi-domain MIDOG++ dataset, and on the TUPAC16 dataset as an out-of-domain case. We show that the H-optimus-0 and Virchow models yielded competitive performance, indicating that the latent spaces of current FMs, all trained on image-level self-supervision, are suitable for direct mitotic figure detection and may be slightly more robust on our out-of-domain test case. All code is made available publicly at https://github.com/DeepMicroscopy/FM4MFdet.
Authors: Shuhan Ye, Hongbin Yu, Chenqi Kong, Pingchuan Ma, Chong Wang, Jun Wan, Qixin Zhang
Abstract: Learned video compression relies on accurate temporal modeling to remove redundancy between adjacent frames. However, most existing codecs infer motion solely from discretely sampled RGB frames, making their estimates vulnerable to fast motion, blur, occlusion, weak texture, low illumination, and abrupt brightness changes. Event cameras asynchronously capture fine-grained intensity changes between RGB timestamps and therefore provide complementary evidence about inter-frame dynamics. We propose ENCORE, an Event-Assisted Complementary Motion Refinement framework for learned video compression. ENCORE first employs Complementary Motion Representation (CMR) to decompose aligned RGB-event features into common and modality-specific motion representations. Spatial Energy and Redundancy-Informed Calibration (SERIC) then identifies event-specific responses that are active and novel relative to RGB, suppresses weak or redundant evidence, and predicts a candidate flow correction. Finally, Energy-Aware Routing (EAR) determines where and how strongly the correction should refine the RGB flow. Events serve solely as an auxiliary modality for motion modeling, while RGB remains the only coding and reconstruction target. Experiments on BS-ERGB, HQ-EVFI, and CED demonstrate consistent gains across datasets and GOP lengths. On BS-ERGB, ENCORE achieves up to 20.80% PSNR-RGB and 22.14% MS-SSIM-RGB BD-rate savings, while retaining clear improvements on the other two datasets.
Authors: MD Wahiduzzaman Khan, Mingshan Jia, Xiaolin Zhang, En Yu, Kaska Musial-Gabrys
Abstract: Creating photorealistic animatable head avatars from a single image remains a fundamental challenge in digital human synthesis. While recent 3D Gaussian Splatting methods have achieved promising results, they rely on external tracking pipelines whose latency is excluded from inference measurements. Furthermore, they adopt unified representations that entangle geometrically distinct facial regions, limiting both expressiveness and rendering fidelity. We propose SpiD (Split and Drive), a single-image Gaussian head avatar framework built on two disentanglement axes. The compute axis internalizes per-frame driving, eliminating external tracking dependency at inference. The feature axis decomposes the avatar into three specialized Gaussian branches, each modeling a geometrically distinct facial domain. Extensive experiments demonstrate consistently strong performance against state-of-the-art methods while achieving the fastest inference speed among all compared methods on a single GPU with the complete driving pipeline included.
Authors: MD Wahiduzzaman Khan, Mingshan Jia, Xiaolin Zhang, En Yu, Kaska Musial-Gabrys
Abstract: Modern face reenactment systems achieve impressive pose and expression transfer using geometry-driven representations. However, they largely ignore tongue dynamics, leading to anatomically inconsistent mouth interiors during speech and expressive motions. We introduce the first framework for cross-identity tongue dynamics transfer in face reenactment. We propose a foundation-model-assisted bootstrapping pipeline that produces a dedicated tongue segmentation model for in-the-wild reenactment without curated annotations. We further introduce a spatially constrained latent masked diffusion model for realistic tongue synthesis, with adaptive mask dilation for seamless mouth boundary transitions. Extensive experiments demonstrate improvements of more than two times over all baselines on every tongue-specific metric. We additionally propose a VLM-based evaluation protocol that replicates expert annotation at scale, confirming perceptual superiority across all ablation variants.
Authors: Henglin Liu, Fangyuan Kong, Jing Wang, Yizhou Lin, Nisha Huang, Chang Liu, Xintao Wang, Pengfei Wan, Kun Gai, Xiu Li
Abstract: Recent advances in preference alignment for diffusion-based video generation, particularly via Direct Preference Optimization (DPO), have significantly improved visual quality. However, temporally sparse artifacts such as motion collapse, object flickering, and color oversaturation remain a major barrier to perceptual realism. Existing methods struggle with these issues due to two key limitations: (1) the preference attribution bottleneck, where offline human annotations are costly and fail to accurately capture learning dynamics, while online reward signals are rollout-aware but often unstable and biased; and (2) temporal credit misallocation, where uniformly applied supervision cannot effectively target the brief segments in which artifacts occur. To address these challenges, we propose concentrated Implicit Preference Optimization (cIPO), a post-training framework for video diffusion models. cIPO derives implicit preference signals directly from the denoising process: given a real video, the model adds forward noise and reconstructs it via iterative denoising, treating the original as the preferred sample and the reconstruction as the dispreferred one. This formulation captures inference-time errors without requiring human annotations or external reward models. Moreover, frame-level discrepancies between original and reconstructed videos reveal when failures occur. cIPO leverages this by computing temporal reconstruction errors and concentrating optimization on high-error segments, enabling more precise correction of failure-prone regions. Extensive experiments demonstrate that cIPO consistently enhances video authenticity and temporal coherence across multiple datasets, highlighting the effectiveness and efficiency of implicit preference with temporally concentrated optimization.
Authors: Sarang Joshi, Peter W. Michor, Stefan Sommer
Abstract: We present a unification of Kendall's landmark shape spaces, where rigid motions are factored out and scale fixed on landmark configurations equipped with Euclidean geometry, with landmark configuration spaces carrying Riemannian metrics descending from right-invariant Sobolev metrics on the diffeomorphism group. The resulting new landmark shape spaces achieve the defining properties of both approaches: The regularity of the descending metric prevents landmarks from colliding, the metric is defined in the ambient space independent of the number of landmarks, local rigid transformations are preserved, global rigid motions are removed, and scale fixed. To achieve this, we define a particular Sobolev-type operator, the screened elasticity operator, whose null-space consists exactly of the rigid motions, we show how this operator descends to achieve the desired geometry, and we present approaches to solving matching problems and computing geodesics numerically. The resulting construction allows the use of landmark configuration spaces with sufficiently regular metrics in applications while retaining the shape invariances that are a hallmark of Kendall's shape spaces.
Authors: Yabin Xu, Fangtao Zhang, Fan Wang, Zhan Wang, Honghua Chen, Mingqiang Wei, Haoran Xie, Sam Kwong
Abstract: Wind turbine blade defect detection remains highly challenging in real-world inspection scenarios due to limited on-site data and the subtle visual characteristics of defects. In practice, blade defects are often small-scale, low-contrast, and difficult to distinguish from complex backgrounds, which significantly limits the robustness of existing detectors. To address these challenges, we propose BladeYOLO, a defect detection framework for wind turbine blades. Specifically, we integrate a Vision Transformer (ViT) backbone initialized with DINOv3 self-supervised pre-trained weights into YOLOv12-L, enabling the transfer of large-scale generic visual priors to blade defect detection and improving feature representation under limited training annotations. To enhance the perception of subtle defects, we further develop a Mamba-guided Weak-Defect Enhancement module, which consists of a Detail-Enhanced Multi-scale Branch for preserving high-frequency structural cues and a Cross-Mamba module for progressively propagating high-level semantic guidance to shallow features. In addition, we introduce a lightweight Style-Injector module that captures environment-related style information via Fourier decomposition and injects it into selected ViT self-attention layers, thereby improving robustness against environment-induced appearance variations. Extensive experiments demonstrate that BladeYOLO achieves superior performance on the WTBlade-Defect dataset, with additional annotation-budget experiments showing its favorable performance under reduced training annotations. Evaluation on the public Wind Surface Defect dataset further provides supportive evidence for the cross-dataset robustness of BladeYOLO. In particular, on this public dataset, BladeYOLO outperforms the best competing method by 3.5\% in mAP$_{50}$ and 2.5\% in mAP$_{50-95}$.
Authors: Jianyi Zhou, Chenghao Zhang, Yanli Li, Dong Yuan
Abstract: Indoor mmWave radar perception is difficult to reproduce because measured range-angle responses depend on scene geometry, material response, multipath, hardware conventions, and signal processing. Existing ray-tracing and digital-twin tools often expose rendering, channel, or path-level quantities, while radar sensing requires complex signal products that can be processed and compared in the same domain as real FMCW measurements. We present mmRadarTwin, a signal-level and path-attributed digital-twin platform for indoor mmWave radar. mmRadarTwin links a real radar measurement branch with an Unreal Engine scene-simulation branch through a shared receive-channel and range-angle processing interface. The simulator writes complex multi-channel receive grids and exports per-path contribution records that identify the actor, material tag, propagation event, and output-bin support of each simulated return. We evaluate mmRadarTwin in an office deployment using a commodity monostatic mmWave radar and mobile scene-capture hardware. Across 154 measured poses spanning 22 radar locations, the current physics-only path-basis simulator recalls 70.8% of measurement-active geometry-supported response regions in the central usable field of view while exposing residuals caused by weak or missing path support, shifted responses, unsupported anchors, and missing physical mechanisms. Rather than claiming complete radar-map reconstruction or cross-room generalization, mmRadarTwin establishes a practical systems workflow for constructing, comparing, and diagnosing indoor radar digital twins.
Authors: Yiheng Xiong, Luisa Gall\'ee, Daniel Santak Wolf, Heiko Hillenhagen, Michael G\"otz
Abstract: Numerous unsupervised domain adaptation (UDA) algori-thms exist, but for clinical practice, selecting the best-suited one along with proper hyperparameters often remains unclear, as the unlabeled deployment (target) domain prevents direct evaluation. We propose a label-free criterion that jointly selects the algorithm and hyperparameters for UDA. Given a pool of candidate models from multiple algorithms trained with different hyperparameters, our approach scores each candidate against an agreement reference, and selects the one with the highest score. The agreement reference is constructed in two levels without using target labels. First, we leverage multiple label-free selection signals, using each to nominate a model within every algorithm. Second, the nominated models are aggregated across algorithms to form a reference prediction for each unlabeled target sample. The candidate whose predictions agree most with this reference is then selected for deployment. Experimental results on four brain MRI and four chest X-ray datasets across seven clinically relevant transfer scenarios show that our method achieves better selection performance than other methods and remains effective across different algorithm pools. Our approach takes a step towards practical, label-free algorithm selection for clinical deployment of UDA.
Authors: Taewan Kim, Jiwoo Kang
Abstract: Face-and-voice association learning is one of the most challenging tasks in deep learning. In this paper, we propose a simple but powerful cross-modal feature embedding method for the association of faces and voices. Previous work has studied cross-modal association tasks to establish the correlation between voice clips and facial images. These works have addressed cross-modal discrimination but underestimate the importance of handling heterogeneity in inter-modal features between audio and video, resulting in a lot of false positives and false negatives. To tackle the problem, the proposed method learns the embeddings of cross-modal features by making another feature exist between cross-modal features, facilitating the voice and face features of the same person to be embedded in a convex hull. Moreover, the incorporation of cross-modal attention mechanisms with convex embedding techniques represents a highly effective strategy for the attenuation of false positives and false negatives, accomplished via the minimization of inter-class discrepancies. We exhaustively evaluated our method for cross-modal verification, matching, and retrieval tasks on the large-scale VoxCeleb dataset. Extensive experimental results demonstrate that the proposed method achieves notable improvements over existing state-of-the-art methods.
Authors: Juheon Hwang, Taewan Kim, Jiwoo Kang
Abstract: We propose a novel collaborative approach for face super-resolution (SR) and robust person re-identification from sequential or multi-view facial images. Traditional SR methods often suffer from blurring and distortion in faces recovered from poor-quality images due to low resolution. Image- and video-based facial SR methods using facial landmarks or segmentation also have similar challenges. To overcome these limitations, we leverage multiple correlated facial observations, across time or viewpoints, by introducing a transformer-based collaborative feature aggregation method that unifies identity features from multi-sequence or multi-view data. This allows faces in multiple sequences of an individual to contribute to accurately estimating common facial features. Furthermore, we propose a cascade SR network to progressively restore the high-resolution image of the target's face with gradual facial feature unification. The unified identity representation is further utilized in person re-identification scenarios, enabling accurate matching even under severe image degradation. The exhaustive experimental results and comparisons show that our method outperforms other state-of-the-art methods, demonstrating consistent improvements in both face super-resolution and re-identification performance. Our work highlights the effectiveness of joint identity reconstruction and progressive image restoration from multiple facial inputs in enhancing downstream visual recognition tasks.
Authors: Juheon Hwang, Taewan Kim, Heeseok Oh, Jiwoo Kang
Abstract: We propose a convolutional neural shading (CNS), a novel pipeline to reconstruct high-quality 3D shapes from multi-view images. Several recent studies have used neural radiance fields and other neural differentiable rendering methods to understand 3D geometry. However, these approaches rely on single-point geometric information, such as positions and normals of the surface, leading to a lack of detailed local geometry. Our approach addresses the inherent limitations of single-point information by leveraging a neural shader to capture variations even in dark and textureless regions with a convolutional neural shader, resulting in far more accurate geometry predictions. Additionally, our method mitigates surface irregularities at image boundaries by introducing a fine-detail displacement network, which utilizes spatial information of surface geometry and learns fine displacement details by correlating neighboring values in the rendering coordinates. Through extensive experiments, our proposed method has demonstrated significant quality improvements in the reconstructed shapes and rendered images over current state-of-the-art methods.
Authors: Longxia Gao, Linan Wang, Yuhe Han, Junze Geng, Meng Zhang, Hanqing Zhao
Abstract: Deep learning has shown promise for automated tongue diagnosis in traditional Chinese medicine (TCM), yet the design space remains underexplored. We conducted a systematic ablation study spanning 20+ model versions under rigorous 5-fold cross-validation on TongueDx2 (5,109 images, 976 expert-annotated) and a merged dataset of 11,101 samples. We compared six backbone architectures, four loss functions, five augmentation strategies, and six training strategies. The best 976-sample model achieved weighted-F1 of 0.6625 using ConvNeXt-Tiny with restrained augmentation and weak-group ensemble, while the best 11,101-sample model reached weighted-F1 of 0.7761. Six key design principles emerged: (1) ConvNeXt-Tiny offers optimal parameter efficiency; (2) BCE substantially outperforms Asymmetric Loss (+2.7%); (3) restrained color augmentation is critical; (4) weak-group ensemble replacement (+2.1%) outperforms probability averaging; (5) data scaling yielded +20.6% improvement; (6) expanding from 13 to 45 label dimensions caused catastrophic collapse (0.78 to 0.22). These principles are generalizable to multi-label medical image classification with class imbalance.
Authors: Shoutai Zhu, Tianyang Xu, Bin Sun, Mingyuan Xu, Yu Liu, Qinzhen Guo
Abstract: Interleaved multimodal Chain-of-Thought (CoT) improves visual reasoning by incorporating auxiliary visual evidence into intermediate reasoning. However, existing approaches remain constrained by externally defined reasoning traces and visual operations, limiting their ability to develop flexible and abstract visual thinking. Reasoning with latent has recently offered a promising direction by internalizing intermediate computation into continuous representations. Nevertheless, existing visual-latent methods mainly supervise latent states through alignment with compressed auxiliary visual features, treating them as proxies for visual observations rather than active reasoning states. Consequently, they capture the provided evidence but fail to fully internalize the abstract reasoning process induced by multimodal CoT. In this paper, we propose OPLD (On-Policy Latent Distillation), a simple framework that transfers the reasoning capability induced by privileged multimodal CoT into latent reasoning representations. Extensive experiments on diverse multimodal benchmarks demonstrate that OPLD consistently outperforms existing latent reasoning methods and achieves state-of-the-art performance on multiple benchmarks. The results suggest that supervising latent representations at the reasoning-process level provides a more effective paradigm for multimodal latent reasoning than conventional feature-level alignment.
Authors: Hail Song, Seokhwan Yang, Jiwon Yang, Woojin Cho, Woontack Woo
Abstract: We propose S-Avatar, a novel method for generating photorealistic 3D head avatars from a single image using a diffusion-guided 3D model generation module and strategies for animating 3D Gaussian Splatting (3DGS). While single-image head avatar reconstruction is crucial for lifelike Virtual Reality (VR) applications, existing approaches often struggle to preserve 3D consistency under unseen viewpoints. S-Avatar addresses this limitation through a three-stage pipeline. First, a high-resolution 3DGS is synthesized directly from a single image using a diffusion-based Gaussian splat generation module. Next, the parametric head model FLAME is aligned with the generated 3DGS by optimizing its parameters and spatial transformations. Finally, to adapt the 3DGS to FLAME variations, we construct a binding template that encodes the spatial relationship between the initial splats and FLAME. The dynamic 3D head avatar can then be rendered in real time by deforming the 3DGS with the binding template. By combining diffusion-guided canonical 3DGS generation with FLAME-based control, our method achieves efficient and accurate reconstruction with enhanced 3D consistency. Evaluations on public datasets demonstrate that S-Avatar outperforms state-of-the-art methods in novel-view and expression generation, achieving superior realism and consistency. Consequently, our approach represents a significant advance in accessible avatar creation, applicable to a wide range of VR/AR applications. The project page is available at https://github.com/hailsong/savatar.
Authors: Haiyang Wu, Weiliang Mu, Zhuofei Du, Dandan Zhong, Kaijie Shi, Haifeng Li, Chao Tao
Abstract: Existing farmland remote sensing image (FRSI) segmentation follows a "Think with Intra-Image" paradigm, assuming that the current image contains sufficient visual evidence for reliable segmentation. Yet farmland appearance varies with phenology and spatial context and is often confused with other land-cover, making instantaneous, local observations inadequate. Thus, segmentation ambiguity stems not only from limited model representation, but more fundamentally from the required spatio-temporal information lying beyond the current image. Based on this insight, we redefine FRSI segmentation from an information bottleneck perspective as a dynamic decision process driven by task-relevant extra spatio-temporal information gain. We further propose FarmSeeker, a dynamic FRSI segmentation agent that identifies ambiguous regions, reasons about their causes, and queries extra spatio-temporal information on demand for accurate segmentation. To evaluate FarmSeeker, we construct GSFS-Bench, the first global-scale, high-resolution FRSI segmentation benchmark that supports reasoning-querying. Experiments show that FarmSeeker achieves more stable segmentation performance than existing methods. The project is publicly available at: https://withoutocean.github.io/FarmSeeker/
Authors: Ioannis Sarridis, Ioannis Kompatsiaris, Symeon Papadopoulos
Abstract: Vision-Language Models (VLMs) such as CLIP are now foundational to multimodal systems, yet their robustness to spurious correlations remains poorly understood at scale. We present the first large-scale empirical study of 194 publicly available VLMs, including 16 model families, covering a wide range of model sizes, 24 training datasets, and three evaluation benchmarks, namely ImageNet (overall performance), CelebA (typical single-attribute bias), and UrbanCars (complex multi-attribute biases). Across these settings, the Spearman correlation between model scale and performance weakens as evaluation shifts from ImageNet ($\rho{=}0.68$) to single-attribute ($\rho{=}0.48$) and further to multi-attribute ($\rho{=}0.05$) bias benchmarks. In contrast, properties of the training data (size and quality) show more consistent relationships with worst-group accuracy across both bias benchmarks. Notably, curated datasets yield improvements of up to 25% over uncurated alternatives at a comparable scale. Finally, the effect of architectural choices (e.g., patch size, image resolution) is highly context-dependent, varying with the nature of the benchmark, including the type of bias and its spatial distribution within images.
Authors: Haoqing Wang, Xingrun Xing, Wei Xia, Ziheng Li, Yehui Tang
Abstract: Agentic vision-language models (VLMs), which interleave textual reasoning with explicit tool calls such as cropping and code-based image manipulation, have emerged as a compelling paradigm for reliable and interpretable multimodal reasoning. However, recent studies have revealed that such models often use tools unfaithfully. Many process images are irrelevant to the question (e.g., the tool crops the wrong region or misses the queried target), yet the call still receives full credit and the model still answers correctly. Such decorative or misaligned tool calls waste computation and reveal that the model leans on prior knowledge or the original image rather than the evidence it retrieves. This may stem from two limitations of prevailing methods: the tool reward fails to distinguish useful from useless calls, and tool feedback carries no signal of usefulness. To this end, we introduce FaithEyes, a multi-agent self-judging framework. Concretely, we use a VLM to judge whether each process image helps answer the question. The judgement is injected into the reasoning context as part of the tool observation to help subsequent reasoning, and meanwhile is used to scale the tool reward by the helpful-tool ratio to suppress reward hacking. To keep judgement available at evaluation and thus ensure train-test consistency, we further design a multi-agent framework where the model itself serves as a subagent to judge the tool calls from main agent, eliminating any dependence on an external model at inference. Training via a two-stage SFT + RL pipeline on adapted open-source data, FaithEyes attains competitive or superior accuracy across visual perception and reasoning benchmarks, while markedly improving tool faithfulness. The homepage is at https://github.com/Mosi-AI/FaithEyes.
Authors: Zexuan Yan, Yuzhou Wu, Yue Ma, Zonghang He, Kaibo Yin, Xiaobing Tu, Yinggui Wang, Jinkui Ren, Xiantao Zhang, Shijian Wang, Jinghong Liu, Linfeng Zhang
Abstract: Egocentric video offers rich manipulation experience for embodied AI, yet collecting diverse egocentric data across scenes, objects, motions, and embodiments remains costly. We present \method, an egocentric world-action simulator that synthesizes controllable, high-quality manipulation videos to expand scarce real-world training data. \method{} builds on a pretrained video generation prior and introduces two geometry-aware conditioning mechanisms. Online Anchored Projective Memory (OAPM) preserves a first-frame 3D scene anchor while periodically refreshing a recent state during autoregressive generation. Action-3D Rotary Position Embedding (A3D-RoPE) encodes end-effector motion with camera-aware 3D rotary coordinates, injecting action geometry into skeleton-to-video cross-attention for precise control. Together, these components improve visual fidelity, geometric stability, and action alignment in long egocentric rollouts. Moreover, augmenting 400 real trajectories with 400 \method-generated trajectories improves out-of-distribution real-robot success from 77\% to 84\% on single-arm tasks and from 53\% to 70\% on dual-arm tasks, demonstrating that the synthesized data substantially improve downstream WAM generalization.
Authors: Iason Tsardanidis, Alkiviadis Koukos, George Choumos, Vasileios Sitokontantinou, Charalampos Kontoes
Abstract: Accurate and scalable parcel-level agricultural monitoring remains challenging because satellite Earth Observation alone provides only an overhead perspective of agricultural parcels, while optical observations are further affected by cloud-induced temporal gaps. This paper presents Space2Ground 2.0, a multi-source framework integrating Sentinel-1 SAR and Sentinel-2 multispectral time series with geo-tagged street-level imagery acquired using vehicle-mounted cameras and shared through the Mapillary platform. A largely automated processing pipeline performs semantic filtering, image quality assessment, viewpoint-based parcel association, and dataset refinement, transforming large volumes of crowdsourced imagery into parcel-linked, analysis-ready data. Applied over Cyprus during the 2022 growing season, the pipeline produced a curated dataset of 46,050 annotated street-level images, selected from an initial collection exceeding 900,000 images and linked with satellite information for 8,581 agricultural parcels. The practical value of the dataset was assessed through parcel-level crop classification experiments using both single- and multi-source observations. The results demonstrate that street-level imagery provides complementary fine-scale visual information that enhances classification when integrated with satellite time series. Overall, Space2Ground 2.0 provides an openly available benchmark dataset and a reproducible methodology for multimodal agricultural monitoring, with potential applications in visual verification, reduced reliance on costly field inspections, and data-driven agricultural policy implementation.
Authors: Jiwen Liu, Shujuan Li, Xiaohan Li, Zijie Meng, Xinyue Liu, Yulong Xu, Yan Zhou, Guoxin Zhang
Abstract: Video re-shooting aims to regenerate videos with controllable camera motion and viewpoint. Existing methods rely on explicit 3D priors, which are limited by reconstruction quality and often perform poorly when synthesizing previously unseen regions, or on paired videos with different camera trajectories, whose scarcity hinders generalization. We revisit video re-shooting through text-driven semantic viewpoint specification, enabling control over shot scale, viewing angle, and first-/third-person perspective. To this end, we propose TARS, a 3D-free video re-shooting paradigm. Timestep-wise sensitivity analysis reveals that camera motion is primarily established during high-noise stages, where coarse spatiotemporal structures are formed. Based on this insight, we introduce self-supervised training to learn camera dynamics and fundamental visual representations without paired re-shooting data or 3D reconstruction. Through data scaling and joint textual-camera conditioning, TARS supports robust camera and viewpoint control, plausibly synthesizing regions beyond the source view under large camera motions while enabling reverse-angle re-shooting and perspective switching. Extensive experiments show that TARS provides more accurate and temporally consistent camera control than prior methods. Project Page: https://ymlinfeng.github.io/TARS.github.io/
Authors: Simone Giano, Lorenzo Severini, Alessandro Galdelli, Adriano Mancini
Abstract: The deployment of Vision-Language Models (VLMs) in critical domains like disaster management requires high-quality multimodal datasets, especially for transferring knowledge via Data-Free Knowledge Distillation (DFKD). However, existing datasets in this domain either entirely lack descriptive text, such as Incidents1M, or suffer from severe text-image semantic misalignment, such as CrisisMMD. In this work, we present a novel methodology to construct and automatically validate a large-scale multimodal dataset for disaster response. Starting from the vision-only Incidents1M, we successfully recovered 100,000 images and generated high-fidelity textual descriptions using two distinct Qwen3.5 architectures: a 4B dense model and a 35B Mixture-of-Experts (MoE) model. To ensure the generated captions provide reliable semantic anchoring for DFKD, we introduce an image-blind LLM-as-a-Judge validation pipeline leveraging Qwen3.5-9B. By intentionally obscuring the original image from the judge, this evaluator accurately simulates the modality gap of the student model during data-free distillation. Our evaluation across 173,179 label pairs demonstrates a high semantic agreement (78.65/100) between the two architectures. Furthermore, the automated evaluation reveals a conservative captioning behaviour, characterized by a high Precision (77.6%) and low Recall (46.0%). This minimizes the false positive noise, while simultaneously exposing underlying human annotation inconsistencies in the original ground truth. This work provides a scalable, LLM-validated multimodal dataset and a reproducible framework to advance cross-modal knowledge distillation.
Authors: Jianjun Chen, Linlin Wang, Lifang Chang, Limin Huo, Shujiang Song, Yanjia Zhao, Mingwei Shao
Abstract: Hyperspectral imaging is widely used in remote sensing and engineering. Therefore, research on its classification methods is crucial. While CNN and Transformer-based methods have advanced, they still face locality constraints and high computational complexity. To address these issues, we propose an innovative hyperspectral image classification model, MSCM-net. Specifically, first of all, a model architecture combining multi-scale CNN and Mamba is proposed. It consists of a multi-scale feature extraction module (MCSE) and multiple stacked Mamba blocks, which integrates the local feature extraction capability of multi-scale CNN and the long sequence modeling advantage of Mamba. Secondly, the proposed MCSE module consists of multi-scale convolution and SENet. Convolution kernels of different scales extract local information with different receptive fields, enhancing the fusion of spatial and spectral information. Meanwhile, the SENet enables the model to automatically learn the importance of each channel in the multi-scale features. Furthermore, we also propose a dual-branch feature aggregation module, which further effectively extracts and integrates the spectral information contained in the central pixel and the spatial information in the surrounding pixels. Our model has undergone numerous experiments on three widely used benchmark datasets. The experimental results show that MSCM-net can achieve advanced classification performance while reducing computational complexity.
Authors: Junrui Zhang, Jiaqi Li, Yiran Wang, Liao Shen, Zhiguo Cao
Abstract: Monocular depth estimation (MDE) faces challenges with non-Lambertian surfaces and adverse weather conditions due to the visual ambiguities inherent in single-image limited information. Existing works address them in isolation via image inpainting or augmentation, yielding limited robustness gains. Language, as a powerful complementary modality to vision, is demonstrated to enhance the visual perception capabilities of vision-language models (VLMs) via detailed long captions. However, prior language-integrated MDE methods fail to fully harness this potential due to short text input with limited information, coarse global text feature learning, and limited language guidance during depth decoding. To address these limitations, we propose CapDepth, a novel framework for robust MDE that leverages guidance from detailed long captions to alleviate visual ambiguities in both challenging scenarios. First, we design a detailed long caption input template that explicitly conveys rich spatial relationships among multiple atom sentences. Second, a dynamic caption encoder is introduced to extract fine-grained depth-relevant text features via progressive masked attention. Finally, we propose a text-adaptive decoder that guides enhanced depth decoding with text features via stable adaptive layer normalization. Extensive experiments validate the efficacy of CapDepth, which outperforms state-of-the-art methods, achieving depth error reductions of 25.0% on non-Lambertian surfaces and 22.0% under adverse weather conditions.
Authors: Edoardo Ragusa, Giovanni Paolo Canuti, Simone Lugani, Rodolfo Zunino, Paolo Gastaldo
Abstract: While depth sensors have the potential to complement RGB data for affordance segmentation in wearable robots, their usage seems to remain underexplored. The paper proposes two approaches: a reformulated version of hardware-aware neural architecture search, endowed with a newly designed search space to integrate depth (D) information into small-sized deep networks, and a dedicated fine-tuning approach, including a preprocessing layer to merge depth information with RGB data and make it compatible with conventional architectures. In both cases, those methods aim to generate solutions that benefit from modern (portable) hardware accelerators and overcome existing tiny-like approaches, which often fail to tackle critical scenarios due to the severe constraints set by the supporting hardware. Extensive experiments on a pair of real-world datasets demonstrate the effectiveness of the proposed method as compared with existing solutions. The approach presented in the paper generates, in most cases, solutions that identify the Pareto optimal front to balance generalization performance and hardware requirements. The paper also describes the supporting prototype, including a Jetson Nano board and a RealSense RGB-D camera. When considering the energy profile of the device, the overall system can attain real-time performances within an energy budget that is compatible with standard batteries, such as those used in smartphones.
Authors: Pouya Ardekhani, Zahra Dehghanian, Morteza Abolghasemi, Hamid R. Rabiee
Abstract: Open vocabulary 3D scene understanding is essential for next-generation interactive systems, empowering users to intuitively query and navigate reconstructed environments using natural language. However, current 3D Gaussian frameworks are often bottlenecked by restrictive multiview capture requirements, costly scene-specific optimization, and the massive memory overhead of storing dense language features. We present a novel, training-free pipeline that fundamentally reimagines this paradigm by explicitly decoupling 3D geometric reconstruction from semantic integration. Given a standard monocular video sequence as input, our method efficiently outputs a compact, highly interpretable, and fully searchable object-level semantic Gaussian map. Rather than entangling heavy language embeddings within the mapping loop, we extract geometry independently and ground semantics through a lightweight, modular post-processing framework. Extensive evaluations on the Replica dataset demonstrate that this decoupled architecture preserves strong rendering fidelity and competitive segmentation accuracy. Crucially, by replacing dense per-Gaussian storage with modular, object-level semantic embeddings, our approach delivers an order-of-magnitude reduction in memory usage compared to SOTA baselines. This provides a highly efficient, scalable, and practical solution for open-vocabulary 3D retrieval and question answering directly from everyday monocular video.
Authors: Mingkang Dong, Muxin Pu, Jie Li, Bohan Guo, Songruo Chen, Bin Ren, Xu Zheng, Chen Zhao, Tianwen Qian, Mohamed Elhoseiny, Yuqian Fu
Abstract: Streaming video understanding requires models to continuously retain useful visual evidence before future questions are known. Existing approaches primarily manage the growing visual context according to token importance, temporal redundancy, or segment-level relevance, but rarely organize evidence around objects that persist and evolve over time. Thus, in this paper, we introduce ObjectStream, a training-free framework that treats latent objects as memory anchors for streaming video understanding. ObjectStream induces spatially coherent latent objects directly from frozen Video-LLM representations, links them across frames into persistent anchors, and maintains their histories under a bounded memory budget, without requiring external object detectors or segmentation models. Built on these anchors, ObjectStream preserves three complementary forms of evidence: persistent object histories, transient object changes, and recent visual context. This design enables existing Video Large Language Models (Video-LLMs) to reason over object identities, interactions, and state changes while leaving the underlying model unchanged. Extensive experiments on online streaming and offline long-video benchmarks demonstrate both effectiveness and efficiency. In online streaming evaluation, ObjectStream improves Qwen2.5-VL-7B by 10.0 points on OVO-Bench Real-Time Visual Perception, while reducing peak GPU mem-ory and TTFT by approximately 50%. On offline long-video benchmarks, it surpasses the full-token baseline while discarding 82.5% of visual tokens. These results highlight latent objects as a practical and effective organizing principle for compact streaming video memory.
Authors: Peiyi Xu, Junpeng Zhang, Guanbin Li, Ronghua Shang, Mingtao Feng, Le Dong, Weisheng Dong, Guangming Shi, Jie Feng
Abstract: Monocular UAV videos provide valuable observations for dynamic reconstruction of complex urban scenes. However, such scenes exhibit pronounced spatiotemporal heterogeneity: different regions follow distinct temporal activity patterns, while the motion states of some dynamic regions may further evolve over time. Although dynamic Gaussian methods based on decomposed shared spatiotemporal feature fields have achieved efficient and accurate reconstruction in object-centric or relatively compact scenes, their commonly adopted fixed plane-wise feature combination mechanisms are less suited to the heterogeneous local dynamics of UAV scenes, often leading to ghosting artifacts and blurred dynamic details. To address this challenge, we propose AdaAnchor4D, an adaptive anchor deformation framework for monocular UAV dynamic scene reconstruction. At its core, Anchor-Conditioned Feature Aggregation (ACFA) adaptively aggregates shared spatiotemporal features using anchor-specific aggregation embeddings and temporal information, allowing different local units to obtain dynamic representations tailored to their local and temporal states. Decoupled Local Geometry Deformation (DLGD) separates anchor-state deformation from local Gaussian geometry deformation, while Density-Adaptive Coordinate Warping (DACW) reparameterizes feature-query coordinates according to the axis-wise anchor distributions, alleviating the mismatch between non-uniform geometric sampling and uniform grid parameterization. Experiments on UAV-Arc4D, VisDrone, and UAVDT show that AdaAnchor4D achieves higher rendering quality than representative dynamic Gaussian methods while maintaining real-time rendering performance. The code will be made publicly available.
Authors: Maame Owusu-Ansah, Kelvin Lee, Dr Vinod Venugopal, Muhammad Moazzam Jawaid, Wenting Duan, James Brown
Abstract: Fractional flow reserve derived from CT angiography (FFR-CT) simulates flow through a patient-specific vessel model, so its accuracy depends on the connectedness of the segmented tree, not only on volumetric overlap: a segmentation can reach high Dice yet sever a bifurcation, dropping the downstream subtree and reversing the treatment decision. Topology-aware losses such as clDice and Skeleton Recall act on the global centreline and can miss localised breaks. We study the Bifurcation Connectedness Score (BCS), which scores connectedness at each ground-truth bifurcation, and soft-BCS, its differentiable training surrogate. BCS captures a property of segmentation quality the standard metrics miss: it responds strongly to breaks in connectedness while staying largely unchanged under connectedness-preserving narrowing. Higher BCS accompanies closer agreement between the FFR-CT decisions a solver makes on predicted versus ground-truth geometry, most clearly in severe disease (OR 2.16, CI [1.23, 4.18]). Both decisions come from the same solver, so this reflects geometric, not clinical, fidelity. In training, soft-BCS and Skeleton Recall recover the same branches but build different trees. Recovering branches and keeping them connected are separable properties, so we recommend reporting a measure of each.
Authors: Jie Ma, Zhike Qiu, Jie Gao, Jiayi Ji, Qian Chen, Xiaoshuai Sun, Rongrong Ji
Abstract: While visual token pruning is essential for efficient Multimodal Large Language Models (MLLMs), existing training-free methods suffer from a critical limitation: they rely on static, instantaneous heuristics to perform irreversible filtering. This approach ignores the hierarchical nature of MLLMs, where token importance often evolves dynamically rather than remaining fixed across layers. Consequently, tokens essential for deep-layer reasoning are often prematurely discarded by shallow-layer estimates. To address this, we propose Trend-aware Pruning, a novel framework that elevates pruning from a local snapshot decision to a temporal trajectory modeling problem. Instead of relying on isolated scores, our method captures the momentum of attention flow. This enables a dynamic rectification mechanism that selectively reactivates "late-blooming" tokens, those initially undervalued but exhibiting rising semantic importance, thereby preventing the loss of critical visual cues. Extensive experiments demonstrate that our approach achieves a superior efficiency-performance trade-off across diverse multimodal tasks. Notably, it reduces visual tokens by over 77.8%, retaining only approximately 23 tokens in the final layer while maintaining competitive performance, offering a robust and reversible solution for high-efficiency multimodal inference.
Authors: Jin Cao, Zian Meng, Kaipeng Zhang
Abstract: We present ShadowDancer, a novel approach to any-action, frame-level control of interactive video world models. The obstacle is representational: existing interfaces either encode an action loosely, leaving how it unfolds for the model to improvise, or encode it exactly through structured signals that serve one family and are hard to acquire, so precise control across diverse dynamics remains impractical. Demonstration videos are the natural remedy, specifying any dynamics frame by frame; yet a video shows its dynamics only through one particular appearance, a single shadow of the underlying dynamics, so actions learned from demonstrations transfer poorly to new scenes. ShadowDancer addresses this with two key innovations: (1) shadow pairs, video pairs that replay the same dynamics under independently resampled appearance, constructed at scale by our Shadow Library, so that a dynamics family becomes controllable exactly when such pairs can be constructed for it; and (2) cross-shadow prediction, which learns actions by predicting one shadow from the other, so that whatever the pairing resamples is discarded by construction and whatever it preserves becomes the action, yielding a unified dynamics representation that drives a block-causal world model. Any demonstrated clip thus becomes a reusable action asset, replayed in new environments without action labels, motion estimators, or fine-tuning. Experiments demonstrate improved action transfer and long action rollout over strong latent-action and interactive world model baselines across diverse dynamics families, with an average blinded win rate of 86% in rollout comparisons. We show video results at https://ShadowDancer-1.github.io
Authors: Isaac Roberts, Petra Bevandic, Alexander Schulz, Barbara Hammer
Abstract: Image similarity underlies many computer vision applications, yet it is often unclear why two images receive a high or low similarity score. Existing explainability methods often rely on gradient-based attribution maps to provide local justifications for similarity. These approaches struggle to provide global insights into what specifically drives similarity in regions of an embedding space, such as texture, shape, or color. We introduce a model- and metric-agnostic framework that explains image similarity using Concept Activation Vectors (CAVs) extracted automatically via Sparse Autoencoders (SAEs). Given a pair of images, we perturb their embeddings along discovered concept directions and measure the resulting change in a chosen similarity function, yielding concept importances. For image pairs, we provide localization with concept attribution maps. We extend this procedure to group-level settings, explaining what drives similarity across a cluster of images rather than a single pair, and further, we introduce Exemplar Retrieval, aiming to recover samples with similar reasons contributing to similarity. Our experiments show that our latent perturbations are more faithful to the underlying data distribution than pixel-space baselines, and that concept importances linearly recover the true similarity score. Qualitative results further confirm the usefulness of our methods in understanding a model's individual and group similarity judgments.
Authors: Weiquan Lin, Yu Deng, Shiyang Liu, Luping Xiao, Xu Tang, Junzhi Yu, Jiaolong Yang, Lei Zhang, Xingyu Chen
Abstract: Hand-object interaction (HOI) modeling remains challenging because it requires joint reasoning about hand articulation, object geometry, contact, semantics, and dynamics under severe visual uncertainty. Foundation models introduce transferable prior knowledge learned from large-scale cross-domain data, offering new ways to address these challenges beyond task-specific data and models. However, the rapidly growing literature remains fragmented, and existing studies typically describe these methods simply as ``using large models'' without systematically characterizing what knowledge is introduced, where it enters the HOI pipeline, or which HOI uncertainty it helps reduce. This survey presents the first systematic review of foundation-model priors for HOI. We organize the literature into six HOI tasks spanning reconstruction and generation. More importantly, we establish a taxonomy of eight foundation-model sub-priors grouped into geometric, semantic, and visual families. Geometric priors encompass shape retrieval, shape reconstruction, and spatial reconstruction; semantic priors include semantic grounding and language reasoning; and visual priors cover visual representation, image generation, and video generation. Based on this taxonomy, we systematically analyze how different priors are represented, injected, and adapted across HOI pipelines and tasks. Beyond how foundation models empower HOI, we further examine how HOI-derived knowledge is used in robot learning, including human-data pretraining, human-to-robot skill transfer, and HOI-to-robot data generation. Finally, we summarize datasets and evaluation protocols, and discuss limitations and future directions toward more generalizable HOI systems. To support long-term progress, we curate a live repository that continuously aggregates emerging methods and benchmarks.
Authors: Ilya Novikov, Svetlana Illarionova, Ruslan Dzharkinov, Maria Smirnova, Ayrat Abdullin, Anna Korotkova, Mariia Ulianova, Dmitrii Shadrin, Evgeny Burnaev
Abstract: Effective flood monitoring is critical for minimizing the impacts of flood disasters on populations and infrastructure. Yet reliable remote sensing across extensive and environmentally diverse regions remains challenging, as most segmentation algorithms lack the generalisation capacity required for large-scale application, while annotated flood data are scarce and unevenly distributed. This study presents an end-to-end multimodal framework for Russian Federation territories sustainable flood monitoring and damage assessment based on synthetic aperture radar data, multispectral imagery, and digital elevation models with their derivatives, forming a 21-channel input. Using a self-collected multimodal dataset covering seven Russian regions, two strategies for water surface detection under limited data conditions were compared: a supervised U-Net++ model and the self-supervised AnySat architecture pre-trained and fine-tuned for the segmentation task. Under the data conditions of this study, supervised learning proved more effective, while the AnySat-based approach offered greater stability and retains advantages for settings where larger unlabelled data or missing modalities at inference are expected. The best flood area predictions were used to estimate flood impact in urban areas in terms of the area affected, material damage, casualties, and ecological and agricultural impact. The estimations were conducted following the official methodology of the Russian Ministry of Emergency Situations. Applied to the 2019 Tulun flood, the obtained results closely matched official assessments, except for material damage, due to the open-source databases usage. The results demonstrate the potential of deep learning and multimodal satellite data integration for scalable, reliable flood monitoring across diverse environmental and data-limited conditions.
Authors: Katy L. Scott, Sejin Kim, Joshua Siraj, Caryn Geady, Matthew Boccalon, Mattea Welch, Mogtaba Alim, Andrew J. Hope, Benjamin Haibe-Kains
Abstract: Radiomics and imaging foundation models promise non-invasive biomarkers of tumour biology, yet predictive signatures may reflect tumour volume or acquisition artifacts rather than meaningful image structure. We introduce READII-2-ROQC, an open-source framework that uses volume-preserving negative controls to assess whether radiomic and deep imaging features capture independent spatial signals. READII-2-ROQC generates voxel-perturbed images across tumour, background and whole-image regions using configurable randomization strategies, then compares feature behaviour and model performance between original and control images. Applied to three public cancer imaging cohorts, the framework processed 3,552 tumour volumes and extracted PyRadiomics and foundation-model features from original images and nine matched controls. Reproducing published survival and HPV-status signatures, we show that multiple models retain performance after spatial structure is destroyed, revealing volume-driven or contextual confounding, whereas others show perturbation-sensitive signal. READII-2-ROQC provides a scalable quality-control strategy for developing interpretable, biologically grounded imaging biomarkers and reproducible radiomics workflows.
Authors: Ping-Kun Chiang, Kun-Ru Wu, Po-han Li, Sandeep Chinchali, Ufuk Topcu, Yu-Chee Tseng
Abstract: Recent advances in large language models (LLMs) and vision-language models (VLMs) have enabled new possibilities for 3D question answering (3D-QA), a key capability for embodied AI and robotic perception. However, most existing methods rely on 3D-specific training or fine-tuning with costly annotations, limiting their scalability and real-world applicability. We present \textbf{ViewMind3D}, a fully training-free and modular framework for 3D spatial reasoning over multi-view observations of a scene without requiring complete 3D reconstruction. The framework decomposes the 3D-QA task into four interpretable components: (1) question-driven multi-view selection, (2) guided visual grounding with language-conditioned object cues, (3) spatial context encoding via a bird's-eye-view (BEV) viewpoint indicator, and (4) structured answer generation through role-based reasoning. This design enables structured, robust, and interpretable reasoning without requiring model tuning. Experimental results on ScanQA and SQA3D show that ViewMind3D achieves competitive performance compared to prior training-free and fine-tuned 3D-LLMs. In particular, our method improves performance on spatially grounded question types, such as ``What'' questions in SQA3D, while maintaining strong overall accuracy (50.8\%) and achieving 73.4 CIDEr on ScanQA. These results demonstrate that effective 3D reasoning can be achieved through modular orchestration of general-purpose LLMs and VLMs for robotic perception in real-world environments.
Authors: Haiyue Zhang, Yi Bin, Xun Jiang, Zeyu Ma, Duo Peng, Guoqing Wang, Yang Yang, Heng Tao Shen
Abstract: Large vision-language models (LVLMs) have achieved significant progress in video understanding, yet understanding long videos remains challenging due to the large number of visual tokens and limited context windows. Visual sampling provides a practical solution by selecting an informative subset of frames. However, existing methods typically either rely on relevance-aware sampling, leading to redundant frame selection and insufficient temporal coverage, or adopt a fixed sampling strategy regardless of query type. In this paper, we propose VisualRouter, a training-free and plug-and-play framework for query-grounded visual sampling. VisualRouter first classifies each query as either global or local and then applies the corresponding sampling strategy. For global queries, it employs a relevance-coverage hybrid strategy that preserves temporal coverage while retaining query-relevant visual evidence. For local queries, it adopts an event-aware frame selection strategy that performs event partitioning, segment-level frame allocation, and intra-event frame selection, jointly balancing relevance, coverage, and diversity with a limited number of input frames. Experiments show that VisualRouter consistently improves multiple LVLMs over uniform sampling, achieving gains of 5.2%, 7.7%, and 11.6% on Video-MME, LongVideoBench, and MLVU with Qwen2.5-VL-7B, and outperforming existing training-free visual sampling methods under the same setting.
Authors: Darsha Udayanga, Pin-Yu Chen, Payel Das, Qiang Ji
Abstract: Detection and localization of AI-tampered images are critical for trustworthy AI, yet modern generative models have made such manipulations increasingly difficult to identify. While traditional binary classifiers can detect image tampering, they lack interpretability and generalization. Vision-Language Models (VLMs) offer a promising alternative due to their strong visual understanding and reasoning capabilities; however, existing approaches typically rely on supervised finetuning with curated explanations rather than exploiting their inherent reasoning capabilities. In this work, we investigate whether VLMs can be trained to reason about AI-generated image edits using reinforcement learning (RL) rather than explicit reasoning supervision. Motivated by the success in Group Relative Policy Optimization (GRPO), an RL technique that incentivizes the model to reason by generating thinking traces prior to giving the final answer, we propose a GRPO-based training framework that utilizes simple accuracy and format rewards. Given an input image, the model produces a structured reasoning trace and predicts whether the image has been tampered with. A lightweight segmentation model is then guided by the reasoning output to generate pixel-level localization masks. Experiments across multiple image manipulation datasets demonstrate that our approach achieves competitive detection and localization performance compared to state-of-the-art image forgery detectors, despite requiring substantially weaker supervision. We introduce effective intersection over union (eff-IoU), a unified metric to jointly evaluate detection and localization. These results suggest that reinforcement learning provides an effective and scalable mechanism for teaching VLMs to reason about AI-generated content.
Authors: Luca de Martino, Federico Aromolo, Federico Nesti, Giorgio Buttazzo
Abstract: Real-time anomaly segmentation is essential for the safety of autonomous systems. Although recent approaches offer high accuracy, their computational cost limits their deployment on embedded hardware. This work presents an efficient and accelerated pipeline designed for both embedded and desktop platforms, targeting the autonomous driving and railway domains. The proposed approach reformulates the Neyman-Pearson scoring stage of PixOOD, a state-of-the-art out-of-distribution detection method, and deploys the full pipeline through hardware-optimized TensorRT compilation, reaching up to 182 FPS on a desktop NVIDIA RTX 4060 GPU and 75 FPS on the NVIDIA Jetson AGX Orin embedded platform, respectively 20x and 18x faster than the original baseline. The achieved results demonstrate that advanced anomaly segmentation can be efficiently deployed for onboard processing in autonomous driving and railway applications.
Authors: Jingwen Yang, Senmao Wang, Luoyao Kang, Runmeng Cui, Keying Zhang, Yunjia Bao, Haifan Gong, Lin Lin, Haiyue Jiang
Abstract: Fine-grained segmentation of auricular structures in CT is challenging because the ear occupies a small image region, cartilage boundaries are highly irregular, and interfaces between cartilage and surrounding soft tissues are often ambiguous. Clinical annotations may also include both composite structures containing cartilage and adjacent skin and their corresponding cartilage-only regions, producing nested and overlapping labels. We propose a world-model-based segmentation framework that enables iterative anatomical reasoning beyond conventional feed-forward prediction. Built on an encoder-decoder architecture, the framework introduces a deterministic recurrent state-space model into the intermediate latent space. Multi-scale encoder features and partially decoded representations are fused to form a structural observation that initializes the latent dynamics. During inference, the model performs a three-step latent rollout without ground-truth guidance. Hierarchical anatomical actions update the recurrent state and progressively refine the latent representation. The resulting latent trajectory is projected back into the decoder and combined with high-resolution features to produce the final segmentation. To learn reliable latent transitions, we introduce a balanced hierarchical action objective that addresses foreground sparsity, missing anatomical groups, and imbalance between add and remove operations. Extensive experiments show that the proposed framework consistently improves segmentation accuracy and reduces HD95 by more than 43% for small, irregular, and overlapping auricular structures in CT. These results demonstrate the effectiveness of latent world-model reasoning for challenging medical image segmentation.
Authors: Tengfei Liu, Yang Shi, Yuran Wang, Xiaohan Zhang, Yuqing Wen, Yuqi Tang, Qixun Wang, Zhuoran Zhang, Xuanyu Zhu, Weihong Lin, Xinlei Yu, Yujie Wei, Xinwei Long, Fengxiang Wang, Xinlong Chen, Yue Ding, Jialu Chen, Haotian Wang, Yuanxing Zhang
Abstract: Existing video captioning models generate natural descriptions of video content but cannot explicitly ground local visual elements to multiple reference images. We introduce multi-reference image-grounded video captioning, a new task requiring factual video descriptions with phrase-level reference grounding, and propose RefCaptioner, a two-stage post-training framework for this task. RefCaptioner combines mixed-data SFT with Hierarchical Coverage-Discounted GRPO to jointly improve reference selection, phrase-level binding, distractor rejection, and cross-reference consistency while preserving general video-captioning ability. To support training, we construct a corpus containing $20,000$ videos and 171,354 reference images. We further introduce MRVBench, a benchmark for evaluating caption factuality and multi-reference grounding on both real-world and AI-generated videos. Experiments show that RefCaptioner achieves the best overall performance among the open-source models while remaining competitive on standard video captioning benchmarks. Human evaluation further confirms that its captions are preferred by annotators and enable more source-faithful video reconstruction with both open-source and proprietary video generators.
Authors: Bowen Liu, Shuning Wang, Xinpeng Ding, Zhiheng Wu, Bodong Du, Xiaomeng Li
Abstract: Long-video understanding commonly compresses videos into a small set of frames or visual tokens for answer generation. Existing compact pipelines focus on retaining relevant visual content as explicit evidence. Yet making evidence available does not ensure that complementary cues across moments are integrated for answering. Our key idea is to organize selected frames into query-relevant cross-frame evidence before generation. We formulate this post-selection stage as a latent evidence interface and instantiate it with GenEvA ($\textbf{Gen}erative$ $Latent$ $\textbf{Ev}idence$ $\textbf{A}ggregation$), a distribution-guided latent evidence aggregation framework. Specifically, GenEvA uses a query-conditioned evidence distribution to focus aggregation on relevant frames, forming compact cross-frame latent evidence from their frame-specific information. Since cross-frame integration is not always needed, the same distribution determines whether to insert this latent complement. Across four benchmarks and two Video-MLLM backbones, GenEvA consistently improves matched-frame baselines. At 8 frames, it raises the four-benchmark LLaVA-Video average by $+5.2$ points and Qwen2.5-VL accuracy on LVBench by $+10.1$ points. These gains require only $0.11\%$--$0.40\%$ average video-token overhead; analyses further show task-aware allocation and benefits from Adaptive Evidence Invocation.
Authors: Cencen Liu (University of Electronic Science and Technology of China), Wen Yin (University of Electronic Science and Technology of China), Dongyang Zhang (University of Electronic Science and Technology of China), Dongmin Li (University of Electronic Science and Technology of China), Shan Zhao (Jiigan Technology), Bing Su (Jiigan Technology), Tao He (University of Electronic Science and Technology of China), Jielei Wang (University of Electronic Science and Technology of China), Guoming Lu (University of Electronic Science and Technology of China)
Abstract: All-in-one image restoration aims to handle diverse degradations within a unified framework. Existing methods commonly encode heterogeneous degradation conditions in a shared latent space, where degradation-related cues and scene content can remain entangled. We characterize the resulting challenge as dual ambiguity: semantic ambiguity in channel-wise modulation and spatial ambiguity in restoration responses, which can lead to content corruption and residual artifacts. To mitigate this issue, we propose DAR-Net, a Dual-Ambiguity Rectification Network for all-in-one image restoration. DAR-Net first introduces a Degradation Archetype Representation (DAR) module to construct a structured degradation state through simplex-constrained archetype mixture modeling. Based on this state, a Semantic Ambiguity Rectification (SeAR) module generates degradation-aware prompts to improve channel-wise conditioning in the decoder. A Spatial Ambiguity Rectification (SpAR) module further regularizes degradation-aware and complementary features toward orthogonal response subspaces, reducing spatial interference between removal and preservation cues. Extensive experiments on standard all-in-one restoration benchmarks show that DAR-Net achieves the best overall performance under both three-degradation and five-degradation settings, improving the average PSNR over the strongest competitor by 0.14 dB and 0.34 dB, respectively; it additionally shows superior performance on CDD-11 and WeatherBench.
Authors: Alex Andonian, Samuel G Rodriques, Andrew D White, Siddharth M Narayanan
Abstract: Chemical structures appear in patents and the scientific literature as images. For programmatic usage, such as indexing in databases or constructing machine learning model training sets, they must be transformed into line notations. The two common forms of this task are translating an image of a single molecule (optical chemical structure recognition - OCSR) and translating a Markush structure that represents a family of molecules. While prior work in the former case is quite mature, Markush structure parsing remains a challenging task. In this work, we treat both tasks as an image-to-text translation problem. We then propose OCSRGlyph, a state-of-the-art OCSR model, improving performance over prior methods by carefully considering stereochemistry. For the Markush task, we introduce MarkushGlyph, a vision-language model that reads the entire Markush structure as an image. This contrasts with prior systems, which often use multiple stages to separately process visual and text input content. Finally, we introduce a new metric for determining the accuracy of Markush structure translations, handling failure modes present in prior metrics.
Authors: Ruman Wang, Hangting Ye
Abstract: Classifying pathological scars from clinical photographs requires distinguishing keloids from hypertrophic scars despite limited expert-labeled data and substantial acquisition variation across hospitals. End-to-end image models remain data-dependent, whereas sending photographs to a hosted vision-language model (VLM) may conflict with local data-governance requirements and yields decisions that are difficult to reproduce and audit. We introduce ScaFE (Scar Feature Engineering), which transfers clinical knowledge from a large language model (LLM) into deterministic, executable feature programs instead of asking the model to diagnose images. A web-enabled LLM retrieves clinical evidence and synthesizes programs that measure visually assessable scar attributes. Candidate programs execute in a restricted local environment, and only aggregate validation statistics and feature-level SHAP summaries are returned for iterative repair and refinement; raw images and patient-level outputs remain local. A lightweight Random Forest then operates on the resulting structured representation. On 600 photographs from three hospitals under leave-one-site-out evaluation, ScaFE achieves 81.0% site-macro balanced accuracy, exceeding the strongest baseline, BiomedCLIP, by 10.0 percentage points. With only 10% of the development data, ScaFE retains 72.0% balanced accuracy and an 11.8-point lead. Iterative refinement also raises the executable-program rate from 66.7% to 95.0%, with verified evidence for 91.7% of the final features. These results show that LLM knowledge can support data-efficient, cross-site medical image classification through local and auditable feature programs rather than direct VLM decisions.
Authors: Yunzhan Fu, Xiangyu Shen, Yifei Sun, Yuhan Chen, Jian Wu, Hongxia Xu
Abstract: Medical image fusion aims to integrate complementary information from diverse imaging modalities to support clinical diagnosis. Existing methods typically apply uniform fusion rules globally, lacking a deep understanding of diagnostic intents and pathological structures. To address these limitations, we propose MIND, a Multimodal Intent-Driven Network via Diffusion Transformers (DiTs) for medical image fusion. Specifically, we utilize BioMedGPT to generate intent-driven fusion texts from source images, guiding the fusion process with pathology-aware diagnostic intents. To combat the loss of 2D spatial continuity caused by 1D sequence flattening in DiTs, we design a Multi-scale Latent Adapter. This module explicitly extracts source image features before serialization, injecting them into the network via strict dimensional alignment to effectively supplement image features. To resolve the semantic shift caused by decoupling image outputs from diagnostic intents, we design a medical semantic consistency loss. This loss ensures deep semantic locking between fused images and fusion texts while maintaining the stability of the underlying physical manifold reconstruction. Comprehensive experiments on the Harvard, BraTS, and GFP datasets reveal that MIND delivers superior fusion quality, significantly improves downstream brain tumor segmentation accuracy, and enables flexible interactive fusion, holding significant promise for intent-driven intelligent clinical decision support systems.
Authors: Simon Roy, Mark Bong, Giovanni Beltrame
Abstract: Operational Earth observation increasingly calls for answering queries such as ``find the image pairs where a new building appeared.'' This means searching an archive of before-and-after (bi-temporal) satellite image pairs and ranking each pair by how well it matches a natural-language description of the change. The component that performs this match, the fusion module that combines the ``before'' and ``after'' views, must be run at query time across many candidate pairs, so its speed largely sets the cost of every search. We present a controlled comparison of how to build that module. Using one fixed image encoder (a frozen CLIP model) and one training recipe for all variants, we evaluate eight designs drawn from three families: attention, state-space models (Mamba), and learned compression (our Temporal Bottleneck Fusion, TBF). Each design is tested on two benchmarks (LEVIR-CC and Dubai-CC) with ten random seeds, so the reported differences are statistically grounded. We outline three findings: first, a training-free two-stage search (a cheap difference model that shortlists candidates, followed by attention fusion that re-ranks them) matches or exceeds full-fusion recall on LEVIR-CC while cutting query cost $10$-$15\times$, with comparable R@1/R@5 on Dubai-CC; second, the linear-time scan of Mamba, attractive on paper, gives no speed benefit at the patch counts typical of vision transformers ($L{=}196$): the scan is limited by memory bandwidth, whereas attention maps cleanly onto parallel hardware; and third, compressing the fused representation (TBF) reduces parameters by $2.3\times$ and latency by $1.6\times$ for a change-only BLEU-1 cost of $0.007$, although more aggressive compression quietly discards change-relevant detail that aggregate metrics fail to reveal.
Authors: Xiao Luo, Mingyang Du, Xin Zhou, Tianrui Feng, Xiwu Chen, Xiaofan Li, Jiangning Zhang, Dingkang Liang
Abstract: High-fidelity 3D generation predominantly relies on scaling model capacity and data, which incurs prohibitive computational costs. This paradigm typically requires learning geometry from scratch and overlooks the rich semantic and structural priors already encapsulated in discriminative 3D foundation models. We contend that leveraging the profound understanding of the 3D world possessed by these discriminative models can significantly reduce generative cost. To this end, we propose ROAD, a framework that reduces the training cost of 3D generation by transferring these rich discriminative priors into diffusion transformers. To address the inherent semantic-structural heterogeneity between generative and discriminative latents, we introduce a reciprocal-objective alignment strategy. This method synergizes Holistic Semantic Condensing to enforce global semantic coherence and Structural Optimal Alignment, which is formulated as a bipartite matching problem to rigorously align microscopic geometric details between disparate latent spaces. The 3D foundation model is only used for training-time supervision of alignment and is not used at inference, incurring no additional inference cost. Compared with the industrial baseline Step1X-3D, the proposed ROAD achieves highly competitive generation performance with only 1.5% of the training data and significantly reduces training costs, effectively reducing the computational overhead of high-fidelity 3D generation. Code is available at https://github.com/H-EmbodVis/ROAD.
Authors: Md. Mehrab Hossain Opi, Robiul Islam Ryad, Md. Umar Faruk
Abstract: Post-training quantization (PTQ) has emerged as an effective solution for deploying Vision Transformers (ViTs) on resource-constrained devices. However, existing PTQ methods typically employ uniform bit-widths across transformer components, overlooking their heterogeneous sensitivity to quantization and leading to inefficient precision allocation. In this paper, we propose {MixFrag, a fragility-guided mixed-precision PTQ framework for Vision Transformers. MixFrag first estimates component-level quantization fragility by measuring the Kullback--Leibler (KL) divergence between full-precision and isolated quantized output distributions using a small calibration set. It then formulates bit allocation as a Multiple-Choice Knapsack Problem (MCKP), enabling adaptive layer-wise precision assignment under a target bit budget. Extensive experiments on ImageNet-1K across multiple Vision Transformer architectures demonstrate that MixFrag achieves competitive classification performance under practical mixed-precision settings. Furthermore, evaluations on COCO object detection and instance segmentation show that MixFrag achieves state-of-the-art performance among existing mixed-precision PTQ methods, improving the previous best method by up to 9.6 AP under the challenging MP3/MP3 setting. Additional analyses validate the proposed fragility metric and demonstrate its strong correlation with the learned bit allocation. These results establish MixFrag as an effective framework for mixed-precision post-training quantization of Vision Transformers.
Authors: Kangning Zhang, Yixing Li, Shuai Shao, Qingyao Li, Zhengxi Lu, Zhiyuan Yao, Jianghao Lin, Wenxiang Jiao, Yuan Lu, Weiwen Liu, Weinan Zhang, Yong Yu
Abstract: Multimodal on-policy distillation (OPD) transfers fine-grained visual knowledge by supervising student-generated trajectories with a privileged-view teacher. Yet its next-token corrections are source-mixed, combining visual signals with linguistic priors and teacher-specific effects. The key challenge is to estimate which corrections are supported by visual evidence, not merely where or how strongly to distill. We introduce Visual Attribution Distillation (VAD), a counterfactual target-reconstruction algorithm that estimates the visually attributable part of a teacher correction. At each student-generated prefix, VAD evaluates the same fixed teacher with the relevant evidence present and removed. The corresponding change in centered log-probabilities defines ut, a signed proxy for the visual evidence direction that estimates how revealing the evidence supports or refutes candidate tokens. VAD projects the original correction onto this proxy to obtain an intervention-aligned component and a proxy-unexplained residual, then reconstructs a student-anchored target from the former. During training, this reconstructed target supplies the primary supervision signal, while the privileged teacher contributes a weak regularizer. Across six fine-grained visual benchmarks at 4B and 9B scales, VAD outperforms direct privileged-view distillation and visual-advantage weighting. Token- level and controlled-target analyses show that the proxy-aligned component is enriched in task-relevant visual corrections and yields stronger target shifts, especially when evidence refutes a mistaken answer. These results support counterfactual target reconstruction as an effective alternative to source-mixed supervision.
Authors: Qixun Wang, Yang Shi, Letian Cheng, Zhuoran Zhang, Yan He, Yuqi Tang, Qi Zhang, Xinlei Yu, Ruizhe Chen, Tianrun Xu, Yuanxing Zhang, Pengfei Wan, Haotian Wang, Xianghua Ying
Abstract: The fundamental goal of agentic visual reasoning is to improve the success rate of multimodal large language models (MLLMs) on complex tasks, rather than merely equipping them with a sophisticated yet inefficient reasoning paradigm. In this work, we rethink agentic visual reasoning through two key dimensions of tool use: Mode Adaptiveness (MA) and Tool Effect (TE). Mode Adaptiveness characterizes whether an MLLM can recognize when tools are truly necessary and invoke them accordingly, thereby avoiding unnecessary computational overhead while improving performance on challenging problems that require tool assistance. Tool Effect characterizes the actual impact of tool use: tools should extend the model's capabilities on problems unsolvable through text-only reasoning, while avoiding additional errors on problems that the model can already solve without tools. We conduct a comprehensive analysis to quantify these two properties and empirically reveal that existing agentic visual reasoning models exhibit limited Mode Adaptiveness, while the gains produced by tool use on hard examples are largely offset by the harm introduced on easy examples that the models can already solve. Motivated by these observations, we propose Beacon, a novel agentic visual reasoning model that achieves stronger overall performance, improved Mode Adaptiveness, and genuine tool-induced performance gains. At the core of Beacon are the Necessity-Aware Adaptive Reward and the Hint-Guided Capability Expansion mechanism in the reinforcement learning stage, which respectively encourage adaptive tool invocation based on task necessity and strengthen the model's tool-use capability on the most challenging problems. Extensive experiments across diverse benchmarks demonstrate the strong overall performance of Beacon and its substantial improvements in both Mode Adaptiveness and Tool Effect.
Authors: Chongjian Ge, Hanwen Jiang, Tianyu Wang, Jiuxiang Gu, Yiran Xu, Ziwen Chen, Shaoteng Liu, Jing Shi, Yicong Hong, Zefan Cai, Hailin Jin, Hao Tan
Abstract: Visual generation increasingly requires high-resolution images, long videos, and multimodal context, making the quadratic cost of full attention prohibitive. We introduce Chimera, a hybrid visual diffusion backbone with a principled scaling recipe. Chimera processes text, image, and video tokens in one raster-ordered stream without positional embeddings. It combines Kimi Delta Attention (KDA) for long-context state tracking with O(N) complexity, interleaved Multi-head Latent Attention (MLA) for direct global interaction, and modality-aware short convolutions for local spatiotemporal context. Sparse Mixture-of-Experts (MoE) layers expand capacity while controlling activated compute. To scale this heterogeneous architecture, we introduce HeteroP, a module-wise scheme that transfers hyperparameters across width and depth according to each tensor's functional fan-in and model depth. HeteroP yields a consistently tuned family used to fit Chinchilla-style compute-optimal laws for activated model size, training-token count, and image-video data ratio. Guided by these laws, we train an 11B-parameter Chimera with 2B activated parameters. Experiments show three results. First, measured by pretraining diffusion loss, the dense backbone is 1.7x as compute-efficient as a matched full-attention Wan-2.1 2B baseline, while the complete system reaches 7.3x. Second, without length-specific fine-tuning, Chimera extrapolates zero-shot from 5-second training clips to 30-second videos, with only 6.5% FID degradation in the last five seconds. Third, the fitted laws show that compute-optimal image pretraining divides compute nearly evenly between activated model size and training-token count, whereas video pretraining modestly favors model size at higher budgets. These results establish a foundation for designing and scaling efficient long-context diffusion architectures.
Authors: Shuyao Shang, Yuqi Wang, Ruopeng Gao, Xu Chen, Tieniu Tan, Lue Fan, Zhaoxiang Zhang
Abstract: We introduce PhiZero, a physical world model built around physical language, a compact discrete representation of world-state transitions. Existing physical world models typically predict future videos directly in pixel space, leaving the underlying world dynamics implicit within high-dimensional visual predictors. Motivated by humans' ability to abstract predictive structure from visual experience and organize it in natural language for explicit reasoning, we learn physical language from in-the-wild videos through self-supervision and use it to explicitly reason about how the physical world evolves. Accordingly, PhiZero adopts a reason-then-render paradigm: it first infers future world evolution as a physical-language sequence and then renders the inferred transitions into videos. Extensive experiments across generation and understanding benchmarks validate the ability of PhiZero to model physically coherent world evolution. We further show its potential for realistic and interactive world modeling, fine-grained action-conditioned simulation, and zero-shot motion transfer.
Authors: Yukang Cao, Haozhe Xie, Beichen Wen, Runmao Yao, Yinghao Liu, Yue Huang, Zhichao Liao, Yunxiang Wang, Haiheng Liu, Xingshun Tian, Dawei Su, Long Zhuo, Dacheng Tao, Xiaogang Wang, Liang Pan, Ziwei Liu
Abstract: Embodied intelligence faces a fundamental data bottleneck. Models must capture how first-person perception, whole-body motion, dexterous manipulation, object state, sound, and touch evolve together as humans pursue goals over time. Existing datasets fragment this experience across viewpoints, modalities, or spatial scales, leaving the full perception-action loop only partially observed. We introduce the Ambient Capture Engine (ACE), a human-centric data engine that transforms real home environments into spatially calibrated, temporally synchronized recording studios. ACE operates at two complementary scales: a table-scale configuration resolves hand-object manipulation, while a room-scale configuration captures whole-body motion, locomotion, and interactions across a furnished home. ACE records egocentric and multi-view exocentric video, full-body and articulated hand motion, object geometry and 6-DoF trajectories, audio, and tactile signals as a unified multisensory stream. Using ACE, we build ACE-Data-0, comprising 150 hours and 17M video frames across 200 task categories, performed by 50 participants in 2 environments, for a total of 75,000 interaction episodes. The dataset spans atomic manipulation, long-horizon chains of household activities, and human-scene interaction, while preserving natural behavioral variation through goal-level rather than step-by-step instructions. We further introduce a hierarchical benchmark that progresses from signals to scene components and then to interactions. Evaluations of state-of-the-art methods expose substantial gaps under contact, occlusion, egomotion, and long temporal horizons. ACE-Data-0 provides synchronized human demonstrations with aligned perceptual, kinematic, and contact supervision, offering a scalable foundation for imitation learning, world models, vision-language-action systems, and embodied AI.
Authors: Yao Xiao, Reuben Tan, Zhen Zhu, Yuqun Wu, Jianfeng Gao, Derek Hoiem
Abstract: Long visual context poses a challenge for vision-language models: performance degrades as the number of distractors grows, and processing all tokens at once is computationally infeasible under GPU memory constraints. We present ReToken, a single learnable embedding trained as an explicit retrieval target that selects a sparse set of query-relevant visual tokens from a pre-filled visual KV cache. Trained on only a small image-QA dataset, ReToken yields consistent gains across image and video benchmarks: on Visual Haystacks it improves Qwen3VL-8B by 13.4 points and InternVL3.5 by 12.4 points (>20% relative), and on LVBench it transfers zero-shot to long video for an 8.0-point gain with Qwen3VL-8B. Thanks to its lightweight design, both training and long-video inference fit on a single H100. Code is available at: https://github.com/avaxiao/ReToken
Authors: Frank Li, Bardia Khosravi, Mohammadreza Chavoshi, Theo Dapamede, YoungSeok Jeon, Janice Newsome, Hari Trivedi, Judy Gichoya
Abstract: Training deep learning models on radiological images requires integrating heterogeneous datasets across different sources, file formats, directory layouts, label schemas, and annotation types. We present RadHarmony, an open-source Python library that provides a unified API for loading, harmonizing, and augmenting radiological datasets, with a primary focus on chest radiographs and early support for computed tomography (CT) and magnetic resonance imaging (MRI). RadHarmony standardizes metadata from 24 public datasets into a single tabular format, wraps MONAI's map-style datasets for deep-learning-ready sample delivery with optional on-disk caching, and supports classification labels, segmentation masks, bounding boxes, and radiology report text through a single interface, with an interactive visualization tool for dataset exploration and verification. To lower the barrier for integrating new datasets, RadHarmony introduces an AI-agent skill that guides the full integration workflow from raw data inspection through code generation and testing. We demonstrate the library's utility by pretraining RadHarmony-ViT, a reference vision transformer baseline that combines three heterogeneous chest radiograph datasets with no dataset-specific code. The code and pretrained model weights are available at https://github.com/f10409/RadHarmony.
Authors: Yogisri Pujitha Chinthoti
Abstract: Automated classification of pulmonary disease from chest radiographs is a widely studied application of machine learning in medical imaging. This paper presents a pilot study evaluating classical texture- and gradient-based feature representations for distinguishing COVID-19 from other forms of pneumonia using the publicly available COVID-19 Image Data Collection (668 posteroanterior/anteroposterior radiographs from 408 patients). Using histogram of oriented gradients (HOG) and gray-level co-occurrence matrix (GLCM) texture descriptors with classical classifiers (logistic regression, random forest, and support vector machine), evaluated under patient-level 5-fold stratified cross-validation to prevent data leakage, we obtain a best mean accuracy of 75.4% and AUC of 0.755, modestly exceeding the 71.6% majority-class baseline. We report these results transparently, including their limitations, and use them to motivate and scope a proposed multi-modal deep learning architecture -- combining convolutional and transformer-based encoders across imaging modalities -- as a direction for future work requiring access to larger, multi-institutional, ethically sourced datasets.
Authors: Supratik Bhowal, Subhrajyoti Basu, Aritra Gir Mahanta, Anik Pal Chowdhury
Abstract: Medical vision-language models (VLMs) generate chain-of-thought (CoT) reasoning before answering clinical questions, but whether this reasoning causally influences predictions remains unclear. We present CoT-Mediate, a behavioral framework that perturbs a single clinically meaningful attribute within a model's own generated reasoning and measures whether the resulting prediction follows the edited reasoning. Our framework combines a dual-arm protocol comparing re-prompted evidence with prefix-forced continuation, together with a provenance-controlled intervention that varies only the attributed source of identical reasoning to disentangle reasoning mediation from sycophancy. We evaluate LLaVA-Med and MedGemma on 1,000 VQA-RAD samples each. Prefix-forced continuation consistently yields higher mediation faithfulness than re-prompting, while the provenance analysis reveals distinct model-specific deference behaviors. Across both models, removing visual evidence increases reliance on injected reasoning, whereas laterality is the least faithfully tracked clinical attribute. These results show that the mechanism used to inject reasoning substantially affects measured faithfulness and that contextual position, rather than stated provenance, is the primary determinant of whether medical VLMs use their generated reasoning.
Authors: Alexi Gladstone, Heng Ji, Yilun Du
Abstract: The deep learning revolution, kicked off by AlexNet, taught us that end-to-end training beats decomposing a problem into hand-designed stages. Generative modeling, however, has remained the exception-despite generative models being remarkably capable, they are still not trained end-to-end. This is because, at its core, generative modeling is about handling distributions with many modes, and existing scalable approaches handle this the same way, by factoring the generation procedure, which prevents end-to-end generation. In this work, we introduce Explorative Modeling, a new paradigm that instead factors the training loop, exploring K candidate matches between model generations and data, and training on the best, so predictions commit to modes rather than blurring them. We find Explorative Models (XMs) useful in two settings. First, increasing exploration adds a third pretraining axis beyond parameters and data for existing generative models-where scaling exploration monotonically improves performance across both continuous and discrete domains (images, video, and language). Notably, gains from exploration increase with scale, climbing from 7% to 36% as data scales and from 13% to 23% as models grow, with efficiency gains more than doubling at 3x the compute. Concretely, exploration improves FLOP efficiency by 4.1x, sample efficiency by 6.2x, parameter efficiency by 47%, lifts the strongest of image-generation recipes to a near-state-of-the-art 1.43 FID on ImageNet without guidance, enables scaling how end-to-end existing models are, and unlocks scaling generalization. Second, XMs enable end-to-end reconstructive generative modeling, matching diffusion on control tasks with 16-256x fewer inference steps. Together, these results establish XMs as both a new pretraining axis for existing generative models and a standalone end-to-end generative modeling paradigm.
Authors: Ajay Sridhar, Jensen Gao, Jonathan Yang, Jean Mercat, Suneel Belkhale, Dorsa Sadigh
Abstract: Recent progress in large-scale imitation learning for robot manipulation has been driven by leveraging datasets across a wide range of robot embodiments. However, achieving significant cross-embodiment transfer is often still challenging. In this work, we study the role of using behavior-aligned representations (e.g., object bounding boxes, language motions, end-effector traces of robot motion) in vision-language-action (VLA) models to promote cross-embodiment transfer. We hypothesize that by possessing invariances across embodiments while being predictive of robot actions, these representations can help unify large-scale cross-embodiment data to enhance transfer. To assess our hypothesis, we develop a simulation-based benchmark designed to assess transfer with diverse cross-embodiment data to new embodiments. Using this benchmark, we compare different representations and ways of incorporating them. We identify that end-effector traces can be particularly beneficial for transfer, representations are generally more useful with larger prior datasets, and can be used to benefit from action-free data. We also demonstrate that they can enhance sim-to-real cross-embodiment transfer, improving task completion progress of real robot policies pre-trained on simulation data by 28%. We provide videos of our evaluations at our website: https://ajaysridhar.com/barx/.
Authors: Rishabh Iyer, Truong Pham, Anay Majee
Abstract: Submodular Information Measures (SIMs) have recently emerged as a powerful framework for representation learning and multimodal learning. In particular, the SCORE framework~\cite{majee2024score} demonstrated that SIMs can serve as effective objectives for supervised contrastive learning. Despite their empirical success, however, the geometric and statistical properties induced by different submodular information measures remain poorly understood. In this work, we develop a unified theoretical framework connecting SIMs to classical concepts in representation learning and statistical pattern recognition. We show that Total Information (TI) objectives characterize intra-class structure: Graph Cut TI recovers within-class variance, LogDet TI recovers generalized variance and covariance volume, and Facility Location TI induces imbalance-aware separation that emphasizes rare and confusable classes. We further show that Mutual Information (MI) objectives capture complementary notions of inter-class structure: Graph Cut MI is closely related to centroid separation and Fisher-style discrimination, LogDet MI captures covariance-aware separation through Mahalanobis distance, and Facility Location MI measures nearest-mode representational overlap. We validate these theoretical characterizations using controlled synthetic experiments that independently vary variance, covariance, class imbalance, class separation, and multimodal overlap. Across all settings, the empirical behavior closely matches the proposed theory. Our results provide the first unified geometric and statistical understanding of submodular information measures and offer principled guidance for selecting and designing SIM-based objectives for representation learning.
Authors: L. Raczynski, W. Krzemien, A. Coussat, M. Bala, B. C. Hiesmayr, K. Klimaszewski, M. Obara, R. Y. Shopa
Abstract: PET provides functional images relying on two-photon coincidences from positron-electron annihilation. In human tissue, about 40\% of annihilations are preceded by Ps formation, of which o-Ps component partially decays into three photons, with the remainder annihilating via pick-off or spin-exchange into two photons. This three-photon channel carries additional information about the surrounding micro-environment, including the three-to-two-photon yield ratio as a potential diagnostic marker. We propose the TRIO algorithm, a novel three-photon event-by-event image reconstruction algorithm formulated as a Bayesian maximum a posteriori inference problem. TRIO unifies time-based trilateration, energy-based reconstruction and, for the first time, a physics-informed prior derived from the QED description of Ps decay within a single probabilistic framework. In contrast to positronium lifetime imaging, which requires a prompt photon and is therefore restricted to specific radionuclides, TRIO relies solely on the three photons and is fully compatible with standard radionuclides such as 18F. Monte Carlo simulation modelled after the Siemens Biograph Quadra scanner demonstrates a mean position error of 1.62~cm, improving by approximately a factor of two over the time-based trilateration (3.05 cm) and by about an order of magnitude over energy-based reconstruction alone (18 cm). More importantly, the proposed Bayesian approach is compatible with existing TOF-PET scanners that can register three-photon annihilation coincidences.
Authors: Shubin Ma, Liang Zhao, Chuanye He, Zhenjiao Liu, Liang Zou, Lin Yuanbo Wu, Yu Shao
Abstract: In recent years, multi-view clustering has attracted widespread research interest. However, due to limitations in data collection devices, data across different views often suffer from misalignment, leading to the partial view alignment problem (PVAP). To mitigate the impact of view asymmetry and irrelevant samples, this paper proposes a framework for partial multi-view clustering via dual alignment and structure enhancement (DAS-PMVC), which leverages view structure consistency and semantic relevance. Specifically, DAS-PMVC includes three parts: \textbf{anchor graph structure alignment}, where sample joint embedding representations with consistent latent space are derived from anchor point relationships for initial view alignment; \textbf{structure-enhanced feature learning}, where the model learns view structure information through pretraining and combines multi-view graph convolutional networks to further extract deep latent features from the aligned graph structure to improve the discriminative power of representations; and \textbf{a dual alignment strategy}, where initial alignment is performed through the anchor graph in the pretraining phase, and contrastive learning loss and the Hungarian algorithm are introduced in the training phase to further optimize the alignment of latent features. Experimental results on various datasets demonstrate that the DAS-PMVC framework outperforms existing state-of-the-art methods in clustering performance, showcasing its effectiveness and superiority.
Authors: Gousia Habib, Laura Ruotsalainen
Abstract: Reconstructing dynamic surgical scenes is crucial for robot-assisted minimally invasive surgery; however, it continues to be difficult because of tissue deformation, occlusions, specular reflections, and restricted viewpoints. In this study, we introduce Endo-NeRF++, a neural rendering framework that accounts for uncertainty in the reconstruction of dynamic surgical scenes. Expanding on EndoNeRF, the suggested approach incorporates multi-resolution hash-grid encoding, temporal feature merging, and uncertainty-informed adaptive sampling to enhance reconstruction accuracy and temporal coherence in deformable endoscopic scenes.The multi-resolution hash-grid representation within the framework effectively captures both coarse and fine anatomical details, while temporal feature blending ensures stable reconstruction during tissue deformation and surgical tool occlusions. Additionally, uncertainty-driven adaptive sampling assigns more samples to uncertain areas to enhance rendering quality and geometric coherence. Experiments on robotic surgical video sequences demonstrate that the proposed uncertainty-guided adaptive sampling improves PSNR by up to 1.22\,dB (4.3\%), increases SSIM by up to 5.3\%, and reduces LPIPS by up to 55.1\% compared with the EndoNeRF baseline.
Authors: Shiwei Gan, Lichen Wang, Xiao Liu, Yafeng Yin, Kuizhuang Liu, Sanglu Lu, Lei Xie
Abstract: Recent advances in sign language (SL) understanding (SLU) have led to remarkable progress in tasks such as continuous SL recognition and SL translation. However, these tasks are designed with predefined objectives, requiring models to learn a fixed mapping from sign videos to glosses or spoken-language sentences. As a result, they provide only a limited assessment of whether a model truly understands the semantic content of SL videos. To address this limitation, \textbf{we first propose a new task, Sign Language Question Answering (SLQA)}, which evaluates SL understanding by requiring models to answer arbitrary natural language questions about SL videos. Unlike previous SLU tasks, SLQA provides a more flexible and comprehensive evaluation framework that assesses multiple reasoning capabilities beyond recognition and translation. To facilitate this task, \textbf{we further construct two SignQA benchmarks} based on PHOENIX14T and CSL-Daily by automatically generating question-answer pairs from existing gloss and sentence annotations using carefully designed templates. The resulting datasets cover five complementary question categories, including position reasoning, structural reasoning, visual search, gloss recognition, and translation understanding. \textbf{Finally, we propose a simple yet effective baseline model} equipped with a Question-Conditioned Modulated Temporal Downsampling module and an in-domain knowledge transfer strategy, enabling effective knowledge transfer from existing SLU tasks while enhancing question-aware temporal feature modeling. Extensive experiments demonstrate that our baseline consistently outperforms representative vision-language models across all question categories, establishing a strong benchmark for future research on SLQA. Datasets are available at:{https://huggingface.co/datasets/hulala/SignQA-2026}.
Authors: Lorenzo Sciandra, Samuele Fonio, Roberto Esposito
Abstract: Understanding and exploiting the training dynamics of overparameterized deep neural networks remains a central challenge in modern machine learning. Recent evidence on Neural Collapse (NC) shows that class representations and classifiers exhibit highly structured geometry, while the Tunnel Effect suggests that only a subset of layers is essential for feature extraction. We combine these two perspectives and propose an NC-inspired training framework for simplifying deep networks during training. Our method monitors representation dynamics through the Inverse Fisher Criterion, a stable and efficient proxy for the variability collapse behavior, to identify both the split point between feature extraction and classification and the training stage at which simplification becomes viable. We then replace the trailing layers with a lightweight classification head and continue training the reduced model. Experiments on image-classification benchmarks across MLP, VGG, and ResNet architectures show that the proposed method achieves substantial parameter reductions while maintaining accuracy comparable to that of the full model. Code to reproduce the experiments can be found at: https://github.com/LorenzoSciandra/NNS.
Authors: Jinpeng Hu, Erqiang Wang, Shan Wang, Zhuo Li, Peipei Song, Xun Yang, Meng Wang
Abstract: Mental health understanding in long-form videos requires nuanced reasoning over observable behavior, interpersonal context, and latent psychological states. Existing benchmarks largely reduce this task to coarse-grained classification, providing limited insight into whether models truly understand psychological phenomena or rely on superficial correlations. To address this limitation, we introduce MMHBench, a comprehensive multimodal benchmark for multi-perspective mental health understanding, comprising 268 long-form videos and 2,184 carefully curated questions. MMHBench organizes the evaluation into two complementary settings: (1) third-person assessment, consisting of 605 questions that focus on the interpretation of observable behaviors and multimodal evidence, and (2) first-person perspective-taking, comprising 1,579 questions that require perspective-conditioned reasoning to identify the interpretation of the mental state supported by the available multimodal evidence. We propose a Multi-Agent Question Generation (MAQG) framework that simulates diverse social roles to synthesize questions from multiple perspectives. The generated questions are refined through multi-role feedback and iterative optimization, followed by expert-guided verification to ensure quality and validity. Extensive evaluation of 22 representative multimodal large language models (MLLMs), spanning both open-source and leading closed-source models, demonstrates that long-form video mental health understanding remains highly challenging.
Authors: Dongxiu Liu, Haoyi Niu, Peng Cheng, Yuan Gao, Xirui Kang, Sangli Teng, Koushil Sreenath, Xianyuan Zhan
Abstract: In the physical world we inhabit, space and time are fundamentally continuous. However, existing machine learning paradigms for world modeling are largely confined to discrete-time prediction, thereby exhibiting significant inefficiency in capturing the dynamics of physical world. We introduce Physical-Time Flow (\textbf{PT-Flow}), a novel approach that learns a continuous latent velocity field operating in physical time. Crucially, the underlying dynamics of sequential data are parameterized by an ordinary differential equation (ODE) embedded in a well-structured representation space. Under this paradigm, the prediction of future can be recast as temporal integration via an ODE solver in the compressed latent space. Building upon PT-Flow, we construct \textbf{ODEWorld}, a continuous-time latent world model that is both efficient and versatile. By extracting time-variant features and enforcing ODE properties on both the dynamical representation space and the latent velocity field, ODEWorld effectively addresses the long-standing representation collapse issue in latent world model literature. This also enables high-quality image reconstruction even after long-horizon prediction. Moreover, its continuous nature allows for arbitrary temporal resolution and even backward prediction, which is impossible for most discrete-time models. Lastly, ODEWorld can provide rich planning-oriented information to facilitate downstream policy learning. Comprehensive experiments demonstrate that ODEWorld successfully reconciles planning-conducive dynamics abstraction with visual realism, excelling in both video generation and robotic control. \href{https://dstate.github.io/odeworld_website/}{Project Website}.
Authors: Farzaneh Seyedshahi, Kai Rakovic, Adalberto Claudio Quiros, John LeQuesne, Ke Yuan
Abstract: Understanding tissue organisation in multiplexed imaging requires modelling both cellular phenotypes and their spatial context. Existing approaches typically rely on handcrafted features, such as marker intensity statistics or cell-type proportions, which often fail to scale or generalise across cohorts with heterogeneous marker panels. We introduce MUL-T, a lightweight transformer framework that reframes tissue architecture as a masked contextual prediction task over discrete cell tokens. By learning contextualised [CLS] embeddings without task-specific supervision, the model captures higher-order cellular interactions while remaining computationally efficient. We evaluate MUL-T on several clinically relevant downstream tasks, including core-level tumour pattern classification, patient-level grading, PD-L1 positivity prediction, and cross-dataset treatment response prediction. Across tasks, MUL-T consistently outperforms classical feature-based baselines and achieves performance comparable to a foundation ViT model, despite substantially fewer parameters and lower training cost.
Authors: Shady Gmira, Evangelos Alexiou, Emmanouil Potetsianakis, Emmanuel Thomas
Abstract: We propose Temporally and Spatially Ordered Gaussians (TSOG), a format for efficient representation of 4D Gaussian Splatting (4DGS) content. TSOG extends the Spatially Ordered Gaussians (SOG) framework to the temporal domain by introducing a timeline attribute and temporal parameterization of geometry and appearance attributes. Similar to SOG, TSOG is a lossy format that assigns each Gaussian a unique index and encodes attribute values as index-aligned image data. TSOG is model-agnostic, extensible, and compatible with both discrete and continuous 4DGS representations. Evaluation using a PLYs sequence and FreeTimeGS as baselines, serving as simplistic and state-of-the-art 4DGS representations respectively, shows file size reductions exceeding 90%, with PSNR differences ranging between -0.42 and +0.85 dB. These results demonstrate substantial file size savings with minimal quality degradation, enabling efficient representation, storage, and delivery of dynamic scenes for next-generation 4D content.
Authors: Yiming Xu, Jihua Kang, Chunsai Du, Qifan Zhang, Wangqiu Zhou, Yiting Wu, Tianqi Li, Qi Song
Abstract: In demanding professional environments and meeting review scenarios, lengthy text often imposes a high cognitive load. To facilitate efficient information communication, transforming verbose text into logically clear diagrams is essential. Scalable Vector Graphics (SVG) provide an effective representation for this purpose due to their editability and resolution independence. However, current research on Text-to-SVG generation remains hindered by three major challenges: (1) the scarcity of datasets for complex, logic-rich diagrams; (2) the absence of explicit layout priors, which leads to chaotic spatial arrangements; and (3) the lack of fine-grained visual feedback to validate rendered outputs and correct aesthetic defects. To address these challenges, at the data level, we introduce DocMeetSVG-100K, a large-scale SVG dataset tailored for document authoring and meeting review scenarios. At the model level, we propose GVR-Coder, a novel framework designed to generate high-quality logical diagrams from lengthy professional texts. Specifically, we adopt a curriculum-driven rejection sampling fine-tuning to progressively enhance the model's capability in modeling complex structures, while explicitly incorporating layout constraint knowledge during training. In addition, we introduce reinforcement learning from dual rendering feedback, a mechanism that provides implicit feedback through reward signals to jointly optimize structural complexity and visual aesthetics. Furthermore, we design a generate-verify-repair agent loop, which improves generation quality through explicit, fine-grained feedback and targeted refinement. Extensive experiments demonstrate that GVR-Coder outperforms competitive baselines and reliably produces logically coherent and visually appealing diagrams. Code and data are available at https://github.com/CurryaNa/GVR-Coder.
Authors: Zheyuan Zhang, Johnson Wu
Abstract: We present ReGenVC, an end-to-end generative video codec that compresses talking-head video to an ultra-low bitrate and decodes it in real time. The encoder reduces a source clip to a compact bitstream -- a neurally compressed first frame, per-frame pose keypoints, and metadata -- totaling about 26 kB for a 77-frame sequence. The decoder is a four-step distilled diffusion transformer that reconstructs the video conditioned on the transmitted pose and reference frame. Compared with x264/x265, ReGenVC reduces the bitrate to roughly one tenth of that required by traditional codecs (about 26 kB vs. 250--280 kB for essentially artifact-free reconstruction); at a matched ultra-low bitrate, conventional codecs collapse into blocking artifacts while ReGenVC stays sharp by exploiting a strong generative prior. The central obstacle to deploying such a codec is decoder latency: multi-step sampling with transformer and VAE components is too slow for interactive use. We make the decoder real-time through four-step distillation and three model-preserving system techniques: (i) eight-GPU unified sequence parallelism (Ulysses & Ring), (ii) a spatially-split VAE, and (iii) a three-stage overlapped pipeline; an analytical timing model characterizes the real-time feasibility region. On an 8-GPU node, the system sustains 24 fps output (972 ms per 25-frame window, within the 1000 ms budget), enabling a live browser stream without observed frame underruns. A hybrid CPU-GPU deployment further runs the encoder on the CPU at 24 fps and offloads the decoder-side one-shot conditioning encoders to the CPU, reducing the per-GPU memory peak from 21.1 GB to about 7.7 GB. To our knowledge, ReGenVC is the first end-to-end generative video codec to combine ultra-low-bitrate encoding with real-time decoding on an 8-GPU system.
Authors: Hui Zhang, Julian Ferchow, Jie Song, Mirko Meboldt
Abstract: Many dexterous manipulation tasks require the object to remain securely held throughout the interaction. From the perspective of hand-object relational motion, such manipulation comprises four canonical skills: grasping, relocation, in-hand rotation, and in-hand translation. Human hands flexibly compose these skills to accomplish complex tasks. Existing approaches, however, model these skills separately with skill-specific action constraints, objectives, or even dedicated hand morphologies, which breaks the compatibility and continuity required for long-horizon composition. In this work, we present a unified framework that models all four skills in a single formulation that shares the same state and action spaces and a common objective structure. This formulation enables straightforward distillation of a single cross-skill policy that performs strongly on every skill, generalizes to unseen objects, stays robust to disturbances, and chains skills seamlessly into long-horizon manipulation. The framework also transfers effectively across different hand morphologies. Overall, our results suggest that different dexterous manipulation skills can be viewed as instantiations of a shared task formulation, revealing the intrinsic consistency across different behaviors.
Authors: Hanzhang Zhou, Panrong Tong, Xu Zhang, Quyu Kong, Chenglin Cai, Tianyu Xia, Gongjie Zhang, Jianan Zhang, Long Li, Long Chen, Lei Wang, Gaole Dai, Pengxiang Li, Liangyu Chen, Yue Wang, Steven Hoi
Abstract: GUI agents have the potential to become a general purpose executor over existing digital devices. To advance them toward real-world use, we envision agents that operate reliably on real devices, execute workflows across platforms, combine GUI interaction with CLI execution, complete long-horizon tasks, proactively initiate useful services, and autonomously improve their capabilities with minimal human effort. Guided by this vision, we present Qwen-UI-Agent, a real-world centric foundation GUI agent spanning mobile, computer-use, web, and DeepSearch environments. Qwen-UI-Agent combines diverse sandbox environments with a large-scale real-device mobile runtime. Its unified action space interleaves GUI operations with CLI execution and generates batched actions in a single model turn. An AutoResearch-style data flywheel uses agents to construct tasks and environments, diagnose failures, and plan subsequent iterations. Online RL supports training on trajectories exceeding 100 turns, with over 10,000 concurrent environments accelerating rollout. A lightweight harness layer supports proactive service initiation and stateful workflows across mobile and computer. Across a broad suite of evaluations, Qwen-UI-Agent sets state-of-the-art performance on mobile-use benchmarks while delivering competitive performance on computer- and browser-use tasks against frontier models, including Opus 4.8, Gemini 3.1 Pro, and GPT-5.6 Sol. On mobile use, it achieves 82.1% on MobileWorld, 92.2% on MobileWorld-Real, and 97.5% on AndroidDaily. On computer use, it achieves 79.5% on OSWorld-Verified and a 40.0% partial-progress score on OSWorld-v2. On browser use and GUI grounding, it achieves 73.6% on WebArena and 81.5% on ScreenSpot-Pro, respectively.
Authors: Jens Lehmann, Andrei Aioanei, Sahar Vahdati
Abstract: ARC-AGI-3 turns abstraction into an interactive problem of skill acquisition. A player must infer an unfamiliar game's rules, hidden state, and goal while maintaining action efficiency because every move counts. We formalize these environments as parameterized rendered deterministic Moore machines and introduce Tycho, a coding-agent system that constructs and uses game-specific models during interaction. Tycho separates actionable observations from intermediate animation, level-completion, and game-over frames. From this structured history, an agent can model, test, plan with, repair, or bypass a free-form executable hypothesis. In one matched public-set run per policy, we compare four orchestration policies on all 25 public games using Claude Opus 4.8 under matched inference budgets. Actor-requested delegation to a model builder obtains the highest observed mean Relative Human Action Efficiency (RHAE), 88.49. With this selected policy, GPT-5.6 Sol and Opus 5 both reach 100.00 RHAE and complete all 183 levels. Their game-balanced first-run human-replay midranks are 98.5 and 100.0. Opus 5 uses 61% fewer scored actions than the aggregate official human baselines. Automatic repair after verification failures produces models that reproduce observed transitions much more accurately, yet reaches only 83.07 RHAE. Transition match indicates whether a simulator reproduces observed dynamics, not whether it has identified the objective or improves the next action. Strong play also requires deciding when to construct, repair, use, or bypass a model. We call this joint problem active abstraction: generating a testable model from costly interaction and deciding when acquiring or using it is worth its cost.
Authors: Zhoushun Yu, Xiaoyu Hu, Xiangyu Xu
Abstract: Latent world models enable efficient planning by predicting future states in a compact representation space, but their performance depends critically on the quality of the learned latent distribution. LeWorldModel (LeWM) regularizes its latents toward an isotropic Gaussian using the Epps-Pulley (EP) objective. We show that the corrective gradients of EP rapidly vanish for isolated tail samples, leaving heavy-tailed deviations insufficiently controlled. To address this limitation, we propose QQWorld, which replaces EP with a quantile-quantile matching objective that directly aligns projected latent samples with rank-matched Gaussian quantiles, thereby maintaining effective corrective gradients in the tails. We further develop cross-batch QQ, which enlarges the effective ranking pool using detached samples from previous batches, and characterize its bias-variance trade-off. Across four control environments, QQWorld effectively improves the average planning success rate of LeWM, while consistently yielding better Gaussian alignment and thinner latent tails.
Authors: V. S. Usatyuk, D. A. Sapozhnikov, S. I. Egorov
Abstract: We introduce Kohn--Sham Spectral Embedding (KSSE), a physics-inspired energy-based model replacing dense CNN classifiers with a sparse-graph spectral embedding evaluated at the Nishimori temperature of an associated Random-Bond Ising Model. By mapping pre-trained features onto quasi-cyclic low-density parity-check graphs and constructing a regularized Laplacian acting as a Kohn--Sham Hamiltonian, we solve $D$ independent channel spectral problems in $\mathcal{O}(N\log N + k^2_{\text{mode}} N)$ time via FFT on circulant blocks (leveraging Pontryagin self-duality of $\mathbb{Z}/p\mathbb{Z}$) and low-order Rayleigh refinement. Graph topology is optimized using \emph{star-domain surgery}: rather than destroying information-carrying codewords by removing frustrated cycles, we construct edge shifts creating local convexity around codewords while bounding residual frustration to $\rho(B_\gamma)\leq 1+\delta$. Multi-scale fractal analysis ($D_2$ spectrum) and fractal learning-rate landscape certifies a landscape transition from rough regimes ($D_2>3$) to star-domain basins ($D_2<1$), enabling Rayleigh refinement with $k_{\text{mode}}=5$ modes. We prove six theoretical results: a generalized Ihara--Bass identity linking belief propagation to the Laplacian; trapping-set eigenvalue correspondence; additive channel separability with an explicit exchange-correlation bound; a surgery theorem bounding frustration with attractor width $\Omega(1/\sqrt{d_{\min}})$; a quasi-stationarity perturbation bound; and a fixed-point convergence theorem. In a transductive protocol on ImageNet-1000 with frozen EfficientNet-B4 features ($D=1792$), KSSE achieves \textbf{88.93\%} Top-1 accuracy using $\approx 21.24$M parameters, outperforming Swin-L (197M, 86.4--87.3\%) and matching ViT-H/14 (632M, 88.0--89.5\%) under standard inductive setups, while reducing model footprint by $10\times$ and $30\times$, respectively.
Authors: Antonio Delgado-Rosa, David Mu\~noz-Valero, Enrique Adrian Villarrubia-Martin, Juan Moreno-Garcia
Abstract: Airborne surveillance from low Earth orbit is hindered by two interconnected bottlenecks: nanosatellites have a limited downlink budget, yet the conventional approach still transmits terabytes of raw imagery to the ground for processing, and open satellite datasets for aircraft are scarce and severely class-imbalanced. These limitations either delay timely decision-making or prevent standard detectors from learning robust representations of rare aircraft classes. In this paper, a workflow that combines on-board inference with generative data augmentation is proposed to address both limitations jointly. Inference is executed on a 6U CubeSat equipped with a low-power edge tensor accelerator, while a diffusion model fine-tuned through low-rank adaptation generates synthetic minority-class imagery. This synthetic output is automatically annotated, pseudo-labelled, by an intermediate detector and merged with classically augmented samples. The results show that the balanced dataset increases global mean average precision from 77.9% to 82.2%, with the minority class rising from F1=0.683 to F1=0.811, and that the quantised detector fits the on-chip memory and projects 25-30 frames per second on orbit. This approach contrasts with the conventional bent-pipe architecture, in which the satellite acts as a passive data collector. Therefore, the computational tests support the proposed workflow as a decision-support tool for real-time, autonomous airborne surveillance from nanosatellites.
Authors: Qiushi Sun, Kanzhi Cheng, Yian Wang, Bowen Yang, Hang Yan, Liheng Chen, Fangzhi Xu, Zichen Ding, Nuo Chen, Jialin Cao, Xingdong Gong, Zehao Li, Kaiming Jin, Xinfeng Yuan, Zhoumianze Liu, Jingyang Gong, Zhangyue Yin, Jiahui Gao, Zhiyong Wu, Tianbao Xie, Jianbing Zhang, Ben Kao, Lingpeng Kong
Abstract: Computer-using agents (CUAs) are advancing rapidly across the digital world. A CUA trajectory records the agent's actions, states, and reasoning. Verifying whether it fulfilled the task instruction is central to CUA evaluation, data curation, and reinforcement learning. Neither human-written verifiers nor human annotators can provide such verification at scale, so the field increasingly turns to vision-language models (VLMs) as judges of CUA trajectories. But a fundamental question has long gone unexamined: are these VLM judges reliable enough? To study it systematically, we introduce OSReward, a realistic, high-quality benchmark that evaluates VLM judges on CUA trajectories. The trajectories come from diverse agent backbones executing human-verified instructions across platforms, then rigorously labeled with ground-truth verdicts through multi-stage human annotation. Building on it, we derive OSReward-Hard, a challenge set concentrating genuinely hard cases, and OSReward-Multi for fine-grained efficiency and alignment scoring. The most comprehensive evaluation of VLM judges to date finds even state-of-the-art models fall short of an ideal judge, sharing a systematic leniency bias that mislabels failed runs as successes. The few reliable enough to trust are too expensive to run at scale, while affordable open models trail far behind. To close this gap, we construct and release OS-Shepherd-100K, an open corpus of reasoning-annotated trajectory judgments for the CUA community. On it, we train OS-Shepherd (9B and 35B), open reward models that supply low-cost, stable, and reliable reward signals, matching commercial judges at 30-60% lower cost than the frontier. Extensive analyses further inform the design of reliable CUA reward at scale. Our code, benchmark, dataset, and model checkpoints are available at https://os-copilot.github.io/OSReward-Home/.
Authors: Junrui Zhang, Jiaqi Li, Yachuan Huang, Yiran Wang, Jinghong Zheng, Liao Shen, Zhiguo Cao
Abstract: In the field of monocular depth estimation (MDE), many models with excellent zero-shot performance in general scenes emerge recently. However, these methods often fail in predicting non-Lambertian surfaces, such as transparent or mirror (ToM) surfaces, due to the unique reflective properties of these regions. Previous methods utilize externally provided ToM masks and aim to obtain correct depth maps through direct in-painting of RGB images. These methods highly depend on the accuracy of additional input masks, and the use of random colors during in-painting makes them insufficiently robust. We are committed to incrementally enabling the baseline model to directly learn the uniqueness of non-Lambertian surface regions for depth estimation through a well-designed training framework. Therefore, we propose non-Lambertian surface regional guidance, which constrains the predictions of MDE model from the gradient domain to enhance its robustness. Noting the significant impact of lighting on this task, we employ the random tone-mapping augmentation during training to ensure the network can predict correct results for varying lighting inputs. Additionally, we propose an optional novel lighting fusion module, which uses Variational Autoencoders to fuse multiple images and obtain the most advantageous input RGB image for depth estimation when multi-exposure images are available. Our method achieves accuracy improvements of 33.39% and 5.21% in zero-shot testing on the Booster and Mirror3D dataset for non-Lambertian surfaces, respectively, compared to the Depth Anything V2. The state-of-the-art performance of 90.75 in delta1.05 within the ToM regions on the TRICKY2024 competition test set demonstrates the effectiveness of our approach.
Authors: Christian Arnold, Andreas K\"upfer
Abstract: When photographers or media professionals compose an image, they make deliberate choices about what to foreground and what to background, shaping how viewers interpret visual content. Yet most quantitative approaches to image analysis overlook this structure and treat detected objects as equally important. We introduce a framework for measuring object prominence-- the relative salience of objects in an image-- as a means to make computational image analysis attentive to the compositional emphasis a curator has built into an image. Drawing on research in cognitive psychology and computer vision, we outline three approaches for estimating object prominence: size and centeredness, inferred depth, and saliency maps. Validating that curator-composed prominence measurably shifts human visual attention in a pre-registered eye-tracking study, we illustrate this framework's benefits in two further applications. First, we demonstrate how weighting features in line with their prominence can enhance the unsupervised ideological scaling of U.S. newspaper images. Second, we examine gendered visual prominence in U.S. presidential campaign ads from 2016 and 2020, showing that Republican candidates depict women less prominently than their Democratic counterparts. Our framework lets researchers analyze image data at scale while remaining attentive to its communicative structure and intent.
Authors: Zuomin Qu, Wei Lu, Xiangyang Luo, Qian Wang, Xiaochun Cao
Abstract: The misuse of deep learning-based facial manipulation poses a serious threat to civil rights. To prevent such fraud at its source, proactive defense methods have been proposed that embed invisible adversarial perturbations into images, disrupting the manipulation process and rendering the forged output unconvincing to observers. However, non-targeted disruption of the output may leave identifiable facial features intact, potentially leading to the stigmatization of individuals. In this work, we propose a universal framework for combating facial manipulation, termed ID-Guard. The framework employs a single forward pass of an encoder-decoder network to generate cross-model transferable adversarial perturbations. We introduce a novel Identity Destruction Module (IDM) to suppress identifiable features in manipulated faces. The perturbation generation is optimized by formulating the disruption of various manipulation types as a multi-task learning problem, with a dynamic weighting strategy designed to enhance cross-model performance. Experimental results show that ID-Guard effectively defends against diverse facial manipulation models while degrading identifiable regions in manipulated images. It also enables disrupted images to evade facial inpainting and facial recognition systems. Moreover, ID-Guard can be seamlessly integrated as a plug-and-play component into other tasks, such as adversarial training. The source code is publicly available at https://github.com/ZOMIN28/ID-Guard.
Authors: Hadi Rezvani, Navid Zarrabi, Ishaan Mehta, Christopher Kolios, Hussein Ali Jaafar, Cheng-Hao Kao, Sajad Saeedi, Nariman Yousefi
Abstract: Plastic pollution presents an escalating global issue, impacting health and environmental systems, with micro- and nanoplastics found across mediums from potable water to air. Traditional methods for studying these contaminants are labor-intensive and time-consuming, necessitating a shift towards more efficient technologies. In response, this paper introduces micro- and nanoplastics (MiNa), a novel and open-source dataset engineered for the automatic detection and classification of micro and nanoplastics using object detection algorithms. The dataset, comprising scanning electron microscopy images simulated under realistic aquatic conditions, categorizes plastics by polymer type across a broad size spectrum. We demonstrate the application of state-of-the-art detection algorithms on MiNa, assessing their effectiveness and identifying the unique challenges and potential of each method. The dataset not only fills a critical gap in available resources for microplastic research but also provides a robust foundation for future advancements in the field.
Authors: Hangtao Zhang, Yichen Wang, Shihui Yan, Chenyu Zhu, Ziqi Zhou, Linshan Hou, Shengshan Hu, Minghui Li, Yanjun Zhang, Leo Yu Zhang
Abstract: Object detection models are vulnerable to backdoor attacks, where attackers poison a small subset of training samples by embedding a predefined trigger to manipulate prediction. Detecting poisoned samples (i.e., those containing triggers) at test time can prevent backdoor activation. However, unlike image classification tasks, the unique characteristics of object detection -- particularly its output of numerous objects -- pose fresh challenges for backdoor detection. The complex attack effects (e.g., "ghost" object emergence or "vanishing" object) further render current defenses fundamentally inadequate. To this end, we design TRAnsformation Consistency Evaluation (TRACE), a brand-new method for detecting poisoned samples at test time in object detection. Our journey begins with two intriguing observations: (1) poisoned samples exhibit significantly more consistent detection results than clean ones across varied backgrounds. (2) clean samples show higher detection consistency when introduced to different focal information. Based on these phenomena, TRACE applies foreground and background transformations to each test sample, then assesses transformation consistency by calculating the variance in objects confidences. TRACE achieves black-box, universal backdoor detection, with extensive experiments showing a 30% improvement in AUROC over state-of-the-art defenses and resistance to adaptive attacks.
Authors: Haosheng Chen, Lian Luo, Mengjingcheng Mo, Zhanjie Wu, Ji Gan, Jiaxu Leng, Xinbo Gao
Abstract: Event cameras, characterized by microsecond temporal resolution and very High Dynamic Range (HDR), emit high-speed event streams for perception tasks. In recent advancements, Graph Neural Networks (GNNs)-based methods show great potential in event perception. However, they typically rely on straightforward pairwise node connectivity in Euclidean space where they struggle to capture long-range dependencies and faithfully characterize the inherent hierarchical structures of event streams. To this end, we propose EHGCN, a dual-space event perception approach that, to the best of our knowledge, is the first to jointly model event streams in Euclidean and hyperbolic spaces. By introducing hyperbolic geometry into event stream perception, EHGCN enables to naturally capture the anisotropic and hierarchical structures of non-uniform, motion-driven event streams. Specifically, we first introduce a distribution-aware event sifting method based on multi-scale voxel grids and Gaussian distribution modeling, retaining discriminative events while attenuating chaotic noise. Then, we present a Markov Random Field (MRF)-optimized motion-aware hyperedge generation scheme, which minimizes a motion consistency energy function to explicitly capture consistent global motion patterns within short time intervals, thereby eliminating cross-target spurious associations and providing critically topological priors while capturing long-range dependencies among events. Finally, we propose a Euclidean-hyperbolic GCN to fuse the retinal events densely aggregated and hierarchically modeled in local Euclidean and global hyperbolic spaces, respectively, to achieve a hybrid event perception. Extensive experimental results on event perception tasks, such as object detection and recognition, show the effectiveness of our approach. Our code will be released for public use at https://github.com/ev-lluo/EHGCN.
Authors: Hongyu Yan, Kunming Luo, Weiyu Li, Kaiyi Zhang, Yixun Liang, Jingwei Huang, Chunchao Guo, Ping Tan
Abstract: Pose stylization, which aims to synthesize stylized content aligning with target poses, serves as a fundamental task across 2D, 3D, and video domains. In the 3D realm, prevailing approaches typically rely on a cascade pipeline: first manipulating the image pose via 2D foundation models and subsequently lifting it into 3D representations. However, this paradigm limits the precision and diversity of the 3d pose stylization. To this end, we propose a novel paradigm for 3D pose stylization that unifies pose stylization and 3D generation within a cohesive framework. This integration minimizes the risk of cumulative errors and enhances the model's efficiency and effectiveness. In addition, diverging from previous works that typically utilize 2D skeleton images as guidance, we directly utilize the 3D skeleton because it can provide a more accurate representation of 3D spatial and topological relationships, which significantly enhances the model's capacity to achieve richer and more precise pose stylization. Moreover, we develop a scalable data engine to construct a large-scale dataset of ''Image-Skeleton-Mesh'' triplets, enabling the model to jointly learn identity preservation and geometric alignment. Extensive experiments demonstrate that PoseMaster significantly outperforms state-of-the-art methods in both qualitative and quantitative metrics. Owing to the strict spatial alignment between the generated 3D meshes and the conditioning skeletons, PoseMaster enables the direct creation of animatable assets when coupled with automated skinning models, highlighting its compelling potential for automated character rigging.
Authors: Hanwen Shen, Jiajie Lu, Yupeng Cao, Xiaonan Yang
Abstract: Recent advances in AI-generated video have shown strong performance on \emph{text-to-video} tasks, particularly for short clips depicting a single scene. However, current models struggle to generate longer videos with coherent scene transitions, primarily because they cannot infer when a transition is needed from the prompt. Most open-source models are trained on datasets consisting of single-scene video clips, which limits their capacity to learn and respond to prompts requiring multiple scenes. Developing scene transition awareness is essential for multi-scene generation, as it allows models to identify and segment videos into distinct clips by accurately detecting transitions. To address this, we propose the \textbf{Transition-Aware Video} (TAV) dataset, which consists of preprocessed video clips with multiple scene transitions. Our experiment shows that post-training on the \textbf{TAV} dataset improves prompt-based scene transition understanding, narrows the gap between required and generated scenes, and maintains image quality.
Authors: Yuyang Liu, Qiuhe Hong, Linlan Huang, Alexandra Gomez-Villa, Dipam Goswami, Tiantian Peng, Xialei Liu, Joost van de Weijer, Yonghong Tian
Abstract: Vision-language models (VLMs), spanning predictive architectures to generative Multimodal Large Language Models (MLLMs), have revolutionized artificial intelligence through powerful cross-modal alignment and zero-shot generalization. However, enabling them to learn continually from non-stationary data remains a major challenge, as their cross-modal alignment and generalization capabilities are particularly vulnerable to catastrophic forgetting. Unlike traditional unimodal continual learning (CL), VLMs face unique challenges such as cross-modal feature drift, parameter interference due to shared architectures, and zero-shot capability erosion. Furthermore, generative MLLMs exhibit a unique "alignment tax," where catastrophic forgetting manifests not merely as factual amnesia, but as a systemic collapse of deep Chain-of-Thought (CoT) reasoning. This survey presents the first comprehensive diagnostic review bridging continual learning across predictive VLMs and generative MLLMs. We systematically deconstruct the aforementioned failure modes and propose a challenge-driven taxonomy comprising four core paradigms: (1) Multi-Modal Replay Strategies addressing explicit and implicit memory drift; (2) Cross-Modal Regularization enforcing topological and geometric alignment; (3) Parameter-Efficient Adaptation utilizing dynamic routing and subspace projections; and the emerging (4) Model Fusion and Decoupling paradigms. We critically analyze the evolution of evaluation protocols, highlighting the essential shift toward dual-track benchmarks (Domain vs. Ability CL). Finally, we chart a roadmap for future research, emphasizing compositional zero-shot learning, embodied AI with sensor fusion, and autonomous agentic ecosystems. All resources are available at: https://github.com/YuyangSunshine/Awesome-Continual-learning-of-Vision-Language-Models
URLs: https://github.com/YuyangSunshine/Awesome-Continual-learning-of-Vision-Language-Models
Authors: Xinyue Liu, Jianyuan Wang, Biao Leng, Shuo Zhang
Abstract: Fully Unsupervised Anomaly Detection (FUAD) addresses the practical scenario where training data is contaminated with unlabeled anomalies. This setting critically challenges conventional Unsupervised Anomaly Detection (UAD) methods, as they tend to misinterpret training anomalies as normal patterns, leading to false negatives. Although filtering anomalies from the training set is a common countermeasure, it inevitably discards valuable data and degrades the model's representation of normality. To overcome this dilemma, we propose an "isolating to harness" strategy, which isolates the influence of anomalies within specialized divisions and then leverages cross-division collaboration to generate robust pseudo supervision. We materialize this idea via a novel Cross-Division Distillation framework based on the widely studied Reverse Distillation paradigm. CDD first partitions the data into divisions with reduced anomaly ratios to train division-specific students. It then aggregates pseudo-normal features generated by each division-specific student for samples from other data divisions to guide a global student towards a robust anomaly-free representation. Experimental results on noisy versions of multiple AD datasets demonstrate that our method achieves significant performance improvements over the baseline. Code is available at https://github.com/hito2448/CDD.
Authors: Aswini Kumar Patra, Anshu Rastogi, Lingaraj Sahoo
Abstract: Plants in their natural habitats endure an array of interacting stresses, both biotic and abiotic, that rarely occur in isolation. Nutrient stress-particularly nitrogen deficiency-becomes even more critical when compounded with drought and weed competition, making it increasingly difficult to distinguish and address its effects. Early detection of nitrogen stress is therefore crucial for protecting plant health and implementing effective management strategies. This study proposes a novel deep learning framework to accurately classify nitrogen stress severity in a combined stress environment. Our model uses a unique blend of four imaging modalities-RGB, multispectral, and two infrared wavelengths-to capture a wide range of physiological plant responses from canopy images. These images, provided as time-series data, document plant health across three levels of nitrogen availability (low, medium, and high) under varying water stress and weed pressures. The core of our approach is a spatio-temporal deep learning pipeline that merges a Convolutional Neural Network (CNN) for extracting spatial features from images with a Long Short-Term Memory (LSTM) network to capture temporal dependencies. We also devised and evaluated a spatial-only CNN pipeline for comparison. Our CNN-LSTM pipeline achieved an impressive accuracy of 98%, impressively surpassing the spatial-only model's 80.45% and other previously reported machine learning method's 76%. These results bring actionable insights based on the power of our CNN-LSTM approach in effectively capturing the subtle and complex interactions between nitrogen deficiency, water stress, and weed pressure. This robust platform offers a promising tool for the timely and proactive identification of nitrogen stress severity, enabling better crop management and improved plant health.
Authors: Samia Mohinta, Daniel Franco-Barranco, Shi Yan Lee, Albert Cardona
Abstract: Behavioural differences across organisms, whether healthy or pathological, are closely tied to the structure of their neural circuits. Yet, the fine-scale synaptic changes that give rise to these variations remain poorly understood, in part due to persistent challenges in detecting synapses reliably and at scale. Volume electron microscopy (EM) offers the resolution required to capture synaptic architecture, but automated detection remains difficult due to sparse annotations, morphological variability, and cross-dataset domain shifts. To address this, we make three key contributions. First, we curate a diverse EM benchmark spanning four datasets across two invertebrate species: adult and larval Drosophila melanogaster, and Megaphragma viggianii (micro-WASP). Second, we propose SimpSyn, a single-stage Residual U-Net trained to predict dual-channel spherical masks around pre- and post-synaptic sites, designed to prioritize training and inference speeds and annotation efficiency over architectural complexity. Third, we benchmark SimpSyn against Buhmann et al.'s Synful [1], a state-of-the-art multi-task model that jointly infers synaptic pairs. Despite its simplicity, SimpSyn consistently outperforms Synful in F1-score across all volumes for synaptic site detection. While generalization across datasets remains limited, SimpSyn achieves competitive performance when trained on the combined cohort. Finally, ablations reveal that simple post-processing strategies - such as local peak detection and distance-based filtering - yield strong performance without complex test-time heuristics. Taken together, our results suggest that lightweight models, when aligned with task structure, offer a practical and scalable solution for synapse detection in large-scale connectomic pipelines.
Authors: Romain Br\'egier, Gu\'enol\'e Fiche, Laura Bravo-S\'anchez, Thomas Lucas, Matthieu Armando, Philippe Weinzaepfel, Gr\'egory Rogez, Fabien Baradel
Abstract: Parametric body models provide the structural basis for many human-centric tasks, yet existing models often rely on costly 3D scans and learned shape spaces that are proprietary and demographically narrow. We introduce Anny, a simple, fully differentiable, and scan-free human body model grounded in anthropometric knowledge from the MakeHuman community. Anny defines a continuous, interpretable shape space, where phenotype parameters (e.g. gender, age, height, weight) control blendshapes spanning a wide range of human forms---across ages (from infants to elders), body types, and proportions. Calibrated using WHO population statistics, Anny provides realistic and demographically grounded human shape variation within a single unified model. We release the Anny body model and its code under the Apache 2.0 license. Thanks to its openness and semantic control, Anny serves as a versatile foundation for 3D human modeling---supporting millimeter-accurate scan fitting, controlled synthetic data generation, and Human Mesh Recovery (HMR). We further introduce Anny-One, a collection of 780k photorealistic images generated with Anny, showing that despite its simplicity, HMR models trained with Anny can match the performance of those trained with scan-based body models.
Authors: Zhanheng Nie, Chenghan Fu, Daoze Zhang, Junxian Wu, Wanxian Guan, Pengjie Wang, Jian Xu, Bo Zheng
Abstract: Recent Multimodal Large Language Models (MLLMs) have significantly advanced e-commerce product understanding. However, 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. It comprises: (1) a Modality-driven Mixture-of-Experts (MoE) 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 release MBE2.0, a co-augmented Multimodal representation Benchmark for E-commerce representation learning and evaluation at https://huggingface.co/datasets/ZHNie/MBE2.0. 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: 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: Yichen Yang, Hong Li, Haodong Zhu, Linin Yang, Guojun Lei, Sheng Xu, Baochang Zhang
Abstract: Existing autoregressive (AR) methods for generating artist-designed meshes struggle to balance global structural consistency with high-fidelity local details, and are susceptible to error accumulation. To address this, we propose PartDiffuser, a novel semi-autoregressive diffusion framework for point-cloud-to-mesh generation. The method first performs semantic segmentation on the mesh and then operates in a "part-wise" manner: it employs autoregression between parts to ensure global topology, while utilizing a parallel discrete diffusion process within each semantic part to precisely reconstruct high-frequency geometric features. PartDiffuser is based on the DiT architecture and introduces a part-aware cross-attention mechanism, using point clouds as hierarchical geometric conditioning to dynamically control the generation process, thereby effectively decoupling the global and local generation tasks. Experiments demonstrate that this method significantly outperforms state-of-the-art (SOTA) models in generating 3D meshes with rich detail, exhibiting exceptional detail representation suitable for real-world applications.
Authors: Kai Shu, Jingyang Jia, Gang Yang, Long Xing, Xun Chen, Aiping Liu
Abstract: Continuous emotional image content generation (C-EICG) is emerging rapidly due to its ability to produce images aligned with both user descriptions and continuous emotional values. However, existing approaches lack emotional feedback from generated images, limiting the control of emotional continuity. Additionally, their simple emotion-text alignment fails to adaptively adjust emotional prompts according to image content, leading to insufficient emotional fidelity. To address these concerns, we propose a novel generation-understanding-feedback reinforcement paradigm (EmoFeedback$^2$) for C-EICG, which exploits the reasoning capability of the fine-tuned large vision-language model (LVLM) to provide reward and textual feedback for generating high-quality images with continuous emotions. Specifically, we introduce an emotion-aware reward feedback strategy, where the LVLM evaluates the emotional values of generated images and computes the reward against target emotions, guiding the reinforcement fine-tuning of the generative model and enhancing the emotional continuity of images. Furthermore, we design a self-promotion textual feedback framework, in which the LVLM iteratively analyzes the emotional content of generated images and adaptively produces refinement suggestions for the next-round prompt, improving the emotional fidelity with fine-grained content. Extensive experimental results demonstrate that our approach effectively generates high-quality images with the desired emotions, outperforming existing state-of-the-art methods on both our custom dataset and public dataset.
Authors: Minh Thong Doi (IMT Nord Europe, CRIStAL), Vincent Itier (IMT Nord Europe, CRIStAL), Jan Butora (CRIStAL), J\'er\'emie Boulanger (CRIStAL), Patrick Bas (CRIStAL)
Abstract: We introduce DinoLizer, a DINOv2-based localizer of manipulated areas in generative inpainting. The model is trained to focus on semantically altered regions by treating reconstructed areas outside the inpainted mask as a separate class, which yields significant improvements w.r.t. the conven- tional approach. We train the model with LORA on the Query and Value of the transformer blocks and simply add 1 linear layer on top of the backbone to predict manipulations on a 14 x 14 patch resolution. Because DINOv2 only accepts fixed- sized images, we use a sliding window approach to aggregate the predictions on larger images. Empirical results show that DinoLizer outperforms state-of-the-art methods on our proposed dataset and SOTA inpainting datasets. Furthermore, it is very robust to JPEG (double) compression. On average, DinoLizer achieves a 20% higher Intersection over Union score compared to the second best model. The code is publicly available here: https://github.com/anonyme610/dinolizer.
Authors: Yuhang Liu, Wenjie Zhao, Xin Wang, Yunhui Guo
Abstract: Selecting an appropriate pre-trained source model is a critical, yet computationally expensive, task in transfer learning. Model Transferability Estimation (MTE) methods address this by providing efficient proxy metrics to rank models without full fine-tuning. In practice, the choice of which MTE metric to use is often ad hoc or guided simply by a metric's average historical performance. However, we observe that the effectiveness of MTE metrics is highly task-dependent and no single metric is universally optimal across all target datasets. To address this gap, we introduce MetaRank, a meta-learning framework for automatic, task-aware MTE metric selection. MetaRank adopts a retrieve-and-rerank cascade. A lightweight retrieval stage first narrows the metric pool using performance observed on similar meta-training datasets. To address the heterogeneous definitions and scales of MTE scores, the reranking stage represents each retrieved metric through the pairwise source-model ordering it induces on the target dataset. A product-kernel regressor combines dataset and ordering similarities to refine the ranking, enabling cross-metric transfer from historical metrics with similar ordering behavior. MetaRank then ranks the retrieved metrics and selects the most appropriate one to guide source-model selection on an unseen target dataset. Extensive experiments across 11 pre-trained models, 13 candidate MTE metrics, and 14 target datasets demonstrate that MetaRank significantly outperforms all compared baselines.
Authors: Chen-Yang Wang, Ge-Peng Ji, Song Shao, Ming-Ming Cheng, Deng-Ping Fan
Abstract: Camouflage relies heavily on context, but current metrics used in camouflaged object segmentation ignore contextual cues. We identify two major drawbacks of these metrics: first, the Dimension Flaw - a predicted foreground map usually contains both pixel labels and probability scores, whereas ground truth provides only one-dimensional binary labels; second, the Range Flaw - these metrics struggle to capture full-range pixel dependencies. Thus, we propose Context-measure, a novel context-aware evaluation paradigm built on a probabilistic pixel correlation framework. It augments the ground truth with pixel-level contextual affinity and builds a perception cycle, achieving greater consistency with human perception. Extensive experiments using four meta-measures show that our Context-measure comprehensively outperforms all widely adopted metrics for camouflaged object segmentation. To our knowledge, this is the first metric designed for camouflaged scenarios. Code is available at https://github.com/pursuitxi/Context-measure.
Authors: Benjamin Beilharz, Thomas S. A. Wallis
Abstract: While deep learning methods have achieved impressive success in many vision benchmarks, it remains difficult to understand and explain the representations and decisions of these models. Though vision models are typically trained on 2D inputs, they are often assumed to develop an implicit representation of the underlying 3D scene (for example, showing tolerance to partial occlusion, or the ability to reason about relative depth). Here, we introduce MRD (metamers rendered differentiably), an approach that uses physically based differentiable rendering to probe vision models' implicit understanding of generative 3D scene properties, by finding 3D scene parameters that are physically different but produce the same model activation (i.e. are model metamers). Unlike previous pixel-based methods for evaluating model representations, these reconstruction results are always grounded in physical scene descriptions. This means we can, for example, probe a model's sensitivity to object shape while holding material and lighting constant. As a proof-of-principle, we assess multiple models in their ability to recover scene parameters of geometry (shape) and bidirectional reflectance distribution function (material). The results show high similarity in model activation between target and optimized scenes, with varying visual results. Qualitatively, these reconstructions help investigate the physical scene attributes to which models are sensitive or invariant. MRD holds promise for advancing our understanding of both computer and human vision by enabling analysis of how physical scene parameters drive changes in model responses.
Authors: Qianchu Liu, Sheng Zhang, Guanghui Qin, Yu Gu, Ying Jin, Sam Preston, Yanbo Xu, Sid Kiblawi, Wen-wai Yim, Timothy Ossowski, Tristan Naumann, Mu Wei, Hoifung Poon
Abstract: Frontier models have demonstrated remarkable capabilities in understanding and reasoning with natural-language text, but they still exhibit major competency gaps in multimodal understanding and reasoning especially in high-value verticals such as biomedicine. Medical imaging report generation is a prominent example. Supervised fine-tuning can substantially improve performance, but they are prone to overfitting to superficial boilerplate patterns. In this paper, we introduce Universal Report Generation (UniRG) as a general framework for medical imaging report generation. By leveraging reinforcement learning as a unifying mechanism to directly optimize for evaluation metrics designed for end applications, UniRG can significantly improve upon supervised fine-tuning and attain durable generalization across diverse institutions and clinical practices. We trained UniRG-CXR on publicly available chest X-ray (CXR) data and conducted a thorough evaluation in CXR report generation with rigorous evaluation scenarios. On the authoritative ReXrank benchmark, UniRG-CXR sets new overall SOTA, outperforming prior state of the art by a wide margin. We release our model at https://huggingface.co/microsoft/UniRG-CXR.
Authors: Jialuo He, Huangxun Chen
Abstract: Visual token reduction is critical for accelerating Vision-Language Models (VLMs), since visual inputs are represented as token sequences that introduce substantial computational overhead in the LLM backbone. However, most pruning pipelines treat efficiency primarily as a token selection problem and retain a fixed visual token budget across inputs, overlooking the substantial variation in image information density. We propose E-AdaPrune, an energy driven adaptive pruning framework that determines an image specific token budget from the singular value spectrum of the visual feature matrix and passes this budget to existing token selectors. By preserving a certain proportion of spectral energy, our method allocates more tokens to information dense scenes while assigning fewer tokens to redundant scenes, without introducing additional learnable parameters. We evaluate E-AdaPrune across four VLM backbones, three token selectors, and nine benchmarks under matched average token budgets. Results show that E-AdaPrune removes a substantial amount of redundant computation from simple cases and converts the saved budget into larger gains on information rich cases. Notably, on SQA$^\mathrm{I}$ with Qwen2.5-VL-3B, E-AdaPrune uses 35.8\% fewer tokens for simple cases with only a 0.52\% relative performance decrease. The saved budget is redirected to hard cases, which receive 52.5\% more tokens and achieve a 1.94\% relative performance improvement.
Authors: Zhangchi Hu, Yifan Zhao, Yansong Peng, Wenzhang Sun, Xiangchen Yin, Jie Chen, Peixi Wu, Hebei Li, Xinghao Wang, Dongsheng Jiang, Xiaoyan Sun
Abstract: We present RiO-DETR: DETR for Real-time Oriented Object Detection, the first real-time oriented detection transformer to the best of our knowledge. Adapting DETR to oriented bounding boxes (OBBs) poses three challenges: semantics-dependent orientation, angle periodicity that breaks standard Euclidean refinement, and an enlarged search space that slows convergence. RiO-DETR resolves these issues with task-native designs while preserving real-time efficiency. First, we propose Content-Driven Angle Estimation by decoupling angle from positional queries, together with Rotation-Rectified Orthogonal Attention to capture complementary cues for reliable orientation. Second, Decoupled Periodic Refinement combines bounded coarse-to-fine updates with a Shortest-Path Periodic Loss for stable learning across angular seams. Third, Oriented Dense O2O injects angular diversity into dense supervision to speed up angle convergence at no extra cost. Extensive experiments on DOTA-1.0, DIOR-R, and FAIR-1M-2.0 demonstrate RiO-DETR establishes a new speed--accuracy trade-off for real-time oriented detection. GitHub Repository: https://github.com/RicePasteM/RiO-DETR.
Authors: Zhenxuan Zhang, Peiyuan Jing, Ruicheng Yuan, Liwei Hu, Anbang Wang, Fanwen Wang, Yinzhe Wu, Kh Tohidul Islam, Zhaolin Chen, Zi Wang, Peter Lally, Guang Yang
Abstract: Low-field to high-field MRI synthesis has emerged as a promising strategy to improve image quality when access to high-field scanners is limited. However, in ultra-low-field settings, the degradation of anatomical detail is spatially heterogeneous: structurally ambiguous regions are more susceptible to unstable high-frequency generation, which may produce anatomically inconsistent textures and boundaries. This issue is particularly problematic when synthesized images are used for downstream quantitative analysis. We therefore study how to make diffusion-based LF-to-HF synthesis more spatially reliable, rather than only sharper on average. To this end, we propose a reliability-guided diffusion framework (ReDiff) with two complementary inference-time mechanisms. First, a reliability-guided sampling strategy attenuates unstable reverse-diffusion updates in regions with weak low-field support. Second, an uncertainty-aware candidate selection scheme aggregates multiple stochastic reconstructions according to spatial consensus and predictive uncertainty. Beyond aggregate image quality, we test whether the uncertainty is itself a usable reliability signal. Experiments on paired 64mT$\rightarrow$3T MRI datasets show that ReDiff attains the lowest LPIPS across three contrasts and two datasets while remaining competitive on PSNR and SSIM, and downstream segmentation analysis indicates better preservation of anatomical structure.
Authors: Chiyuan He, Zihuan Qiu, Fanman Meng, Runtong Zhang, Linfeng Xu, Qingbo Wu, Hongliang Li
Abstract: Continual learning of pretrained vision-language models (VLMs) is prone to catastrophic forgetting, yet current approaches adapt to new tasks without explicitly preserving the cross-modal semantic geometry inherited from pretraining and previous stages, allowing new-task supervision to induce geometric distortion. We observe that the most pronounced drift tends to concentrate in vulnerable neighborhoods near the old-new semantic interface, where shared visual patterns are easily re-explained by new textual semantics. To address this under an exemplar-free constraint, we propose Semantic Geometry Preservation for Continual Learning (SeGP-CL). SeGP-CL first probes the drift-prone region by constructing a compact set of adversarial anchors with dual-targeted projected gradient descent (DPGD), which drives selected new-task seeds toward old-class semantics while remaining faithful in raw visual space. During training, we preserve cross-modal structure by anchor-guided cross-modal geometry distillation (ACGD), and stabilize the textual reference frame across tasks via a lightweight text semantic-geometry regularization (TSGR). After training, we estimate anchor-induced raw-space drift to transfer old visual prototypes and perform dual-path inference by fusing cross-modal and visual cues. Extensive experiments on five continual learning benchmarks demonstrate that SeGP-CL consistently improves stability and forward transfer, achieving state-of-the-art performance while better preserving semantic geometry of VLMs. Code is available at: https://github.com/chiyuan-IVIPLab/SeGP-CL.
Authors: Weile Gong, Zijian Lu, Mingcai Chen, Yiping Zuo, Xin He, Weibei Fan
Abstract: Vision-language models (VLMs) enable flexible generative optical character recognition (OCR), while their open-ended decoders can expose wrong but fluent text with weak visual support. In audit-sensitive records, such an output can be more costly than abstention. Frozen or externally served VLMs therefore require an external decision layer that can determine whether a transcription has sufficient visual evidence for release. We introduce the Geometric Risk Controller (GRC), a model-agnostic controller that treats controlled geometric transformations as repeatable black-box probes, screens structurally implausible continuations, and releases the unique candidate supported by coherent cross-view evidence. The protocol provides empirical selective exposure control with explicit coverage and query cost under a reproducible fixed decision rule. Experiments across frozen VLMs and standard scene-text benchmarks consistently reduce mean, upper-tail, and catastrophic error among released outputs while retaining high coverage.
Authors: Wenli Zhang, Xianglong Shi, Sirui Zhao, Xinqi Chen, Guo Cheng, Yifan Xu, Tong Xu, Yong Liao
Abstract: Diffusion-based audio-driven talking-head generation enables realistic portrait animation, but also introduces risks of misuse, such as fraud and misinformation. Existing protection methods are largely limited to a single modality, and neither image-only nor audio-only attacks can effectively suppress speech-driven facial dynamics. To address this gap, we propose SyncBreaker, a stage-aware multimodal protection framework that jointly perturbs portrait and audio inputs under modality-specific perceptual constraints. Our key contributions are twofold. First, for the image stream, we introduce nullifying supervision with Multi-Interval Sampling (MIS) across diffusion stages to steer the generation toward the static reference portrait by aggregating guidance from multiple denoising intervals. Second, for the audio stream, we propose Cross-Attention Fooling (CAF), which suppresses interval-specific audio-conditioned cross-attention responses. Both streams are optimized independently and combined at inference time to enable flexible deployment. We evaluate SyncBreaker in a white-box proactive protection setting. Extensive experiments demonstrate that SyncBreaker more effectively degrades lip synchronization and facial dynamics than strong single-modality baselines, while preserving input perceptual quality and remaining robust under purification. Code: https://github.com/kitty384/SyncBreaker.
Authors: Enyi Shi, Fei Shen, Chuancheng Shi, Shuyi Miao, Linxia Zhu, Pengyang Shao, Jinhui Tang, Tat-Seng Chua
Abstract: With the widespread deployment of vision-language large models (VLLMs), their safety alignment faces dual challenges across languages and modalities. Existing methods model multilingual and multimodal safety separately, overlooking coupled risks between low-resource-language instructions and visual contexts, which hinders the detection of cross-lingual and cross-modal harmful intent and the formation of robust safety boundaries. To address this, we propose a neuron-level interpretable safety alignment framework that identifies safety neurons and performs neuron-targeted safety tuning to jointly mitigate multilingual and multimodal risks. Specifically, we compare FFN representations elicited by harmful requests and benign inputs to identify neuron activation strengths associated with safety refusals. Next, we jointly model neuron activations and corresponding down-projection columns to derive neuron-level saliency, separating general multilingual and multimodal neurons from safety neurons responsible for model defense. Finally, neuron-targeted gradient masking restricts parameter updates to the safety subspace spanned by the identified neurons, enabling precise and interpretable safety enhancement. Extensive experiments show that our method enhances multilingual and multimodal safety by tuning only a few safety neurons, while preserving general capabilities.
Authors: Nicolae Cudlenco, Mihai Masala, Marius Leordeanu
Abstract: Authoring a multi-actor scenario for a living 3D world, where every action changes its state, and each action's validity depends on the state accumulated before it, demands the freedom of storytelling and the rigor of simulation at once. We author such scenarios with LLM agents, as Graphs of Events in Space and Time (GESTs) that a simulation engine executes deterministically into narrative videos with per-frame spatial, temporal, and semantic ground truth. A staged pipeline driving a flagship LLM, the standard design in video generation, failed outright: the model violates rules stated verbatim in its prompt, and cannot track the dynamic world state. We answer with a constraint-enforcing tool layer: our Director and Scene Builder agents explore the world's capabilities page by page and build every scene through operations checked against simulator state, so every specification they emit is valid by construction. Because we generate each seed text from an existing scenario graph, we can measure reconstruction: the agent authors its own graph from the text alone, yet matches the original at 0.83 F1 on its events, each action with its participants (0.55 for a random scenario of the same kind), and 0.77 on their ordering (0.43 random). End to end: the standard staged pipeline produced 0 executable specifications in 50 attempts; our agents, driving a budget model, execute 20 of 25 (80%), and are, to our knowledge, the first to exercise the full expressive capacity of GEST.
Authors: Shadid Yousuf, S. M. Mahbubur Rahman, Mohammed Imamul Hassan Bhuiyan
Abstract: The deployment of artificial intelligence in medical imaging is hindered by high computational complexity and resource-intensive processing of volumetric data. Although chest computed tomography (CT) volumes offer richer diagnostic information than projection radiography, their use in AI-based diagnosis remains limited due to the computational burden of processing uncompressed volumetric images (typically stored in NIfTI or DICOM format). Addressing the growing need for low-resource deployment and efficient electronic data transfer, we investigate the utilization of JPEG-compressed chest CT volumes for thoracic abnormality detection. We propose Feature Attention Style Transfer (FAST), a novel distillation framework that transfers both activation patterns and structural relationships from high-fidelity CT representations to a spatiotemporal visual encoder operating on compressed inputs. By combining Gram-matrix-based attention style preservation with dual-attention feature alignment, FAST enables robust feature extraction from degraded volumes. Furthermore, we introduce Structured Factorized Projection (SFP), leveraging Block Tensor Train decomposition as a parameter-efficient alternative to dense projection layers, reducing projection-head parameters by almost half. Our contrastive learning pipeline, CT-Lite, integrates these components with a SigLIP-based multimodal alignment objective. Experiments on CT-RATE, NIDCH, and Rad-ChestCT demonstrate that CT-Lite achieves AUROC within 5-7\% of the uncompressed-input baseline across all three datasets, despite operating on compressed inputs with significantly fewer parameters, paving the way for AI-based clinical evaluation under resource constraints.
Authors: Yue Yu, Jiayu Wang, Jiajia Shi, Zhiyu Tan, Hao Li, Jingjing Chen
Abstract: Makeup transfer aims to apply the makeup style of a reference portrait to a source portrait while preserving identity and background. Early methods formulate this task as unsupervised image-to-image translation, relying on surrogate objectives and often yielding limited performance. Recent diffusion- and flow-based approaches instead exploit synthetic data for supervised training, leading to significant improvements. However, these methods still face two critical challenges: synthetic supervision frequently fails to faithfully preserve identity, and the domain gap between synthetic and real data limits generalization, resulting in degraded performance in complex real-world scenarios. To address these issues, this paper first proposes ConsistentBeauty, a novel data curation pipeline that ensures makeup fidelity and strict identity consistency within the synthesized data. Second, we propose RealBeauty, a synthetic-to-real post-training framework. Beyond supervised learning on curated synthetic data, we further adapt the model to real-world scenarios through reinforcement learning and design novel verifiable rewards tailored to the makeup transfer task. It allows the model to further benefit from real makeup patterns beyond synthetic supervision. In addition, we establish a new diverse benchmark for makeup transfer, covering a wide range of skin tones, ages, genders, poses, and makeup styles, thereby enabling a more comprehensive evaluation of model performance under diverse real-world conditions. Extensive experiments show that our method achieves state-of-the-art performance on multiple benchmarks and demonstrates clear advantages in identity preservation and performance on complex real-world cases.
Authors: Meisen Wang, Hao Deng, Wei Bao, Chengjie Wang, Zhiqiang Tian, Shaoyi Du, Siqi Li, Yue Gao
Abstract: Event cameras provide microsecond-level temporal resolution, low latency, and high dynamic range, offering potential for perception under fast motion and challenging illumination conditions. However, existing Event-based Object Detection (EOD) methods face limitations at both the representation and model levels: prior event representations usually encode temporal information indirectly through redundant structures, while detection models struggle to explicitly aggregate fragmented event responses into coherent high-order object features. To address these limitations, we present \textbf{Event Dual Temporal-Relational Aggregation Detector (Ev-DTAD)}, a unified EOD framework that integrates representation-level temporal encoding with model-level temporal-hypergraph reasoning. Specifically, we introduce \textbf{Hierarchical Temporal Aggregation (HTA)}, a compact three-channel pseudo-RGB representation that explicitly embeds temporal information across intra- and inter-window events. To further enhance detection under sparse and fragmented event responses, we propose \textbf{Frequency-aware Hypergraph Temporal Fusion (FHTF)}, which refines multi-scale event features through temporal evolution modeling and high-order relational reasoning. Extensive experiments on Gen1 (\textbf{+0.8 mAP}), 1Mpx/Gen4 (\textbf{+0.5 mAP}), and eTraM (\textbf{+3.0 mAP}) demonstrate that Ev-DTAD achieves a competitive accuracy--efficiency trade-off, validating the complementarity between compact temporal representation and temporal-hypergraph feature reasoning (Fig.~\ref{fig:bubble}). The code is available at: https://github.com/meisenwang/Ev-DTAD.
Authors: Fuhao Li, Shaofeng You, Jiagao Hu, Yu Liu, Yuxuan Chen, Zepeng Wang, Fei Wang, Daiguo Zhou, Jian Luan
Abstract: Evaluating object removal in images and videos remains challenging because the task is inherently one-to-many, yet existing metrics frequently disagree with human perception. Full-reference metrics reward copy-paste behaviors over genuine erasure; no-reference metrics suffer from systematic biases such as favoring blurry results; and global temporal metrics are insensitive to localized artifacts within edited regions. To address these limitations, we propose RC (Removal Coherence), a pair of perception-aligned metrics: RC-S, which measures spatial coherence via sliding-window feature comparison between masked and background regions, and RC-T, which measures temporal consistency via distribution tracking within shared restored regions across adjacent frames. To validate RC and support community benchmarking, we further introduce PROVE-Bench, a two-tier real-world benchmark comprising PROVE-M, an 80-video paired dataset with motion augmentation, and PROVE-H, a 100-video challenging subset without ground truth. Together, RC metrics and PROVE-Bench form the PROVE (Perceptual RemOVal cohErence) evaluation framework for visual media. Experiments across diverse image and video benchmarks demonstrate that RC achieves substantially stronger alignment with human judgments than existing evaluation protocols. Project page: https://xiaomi-research.github.io/prove/.
Authors: Jiusong Ge, Yingkang Zhan, Wenjie Zhao, Di Zhang, Ke Wang, Jiashuai Liu, Chunze Yang, Chengzu Li, Jian Zhang, Yuxin Dong, Ni Zhang, Qidong Liu, Mireia Crispin-Ortuzar, Huazhu Fu, Chen Li, Zeyu Gao
Abstract: Traditional whole slide image (WSI) analysis methods typically rely on the multiple instance learning (MIL) paradigm, which extracts patch-level features at high magnification and aggregates them for slide-level prediction. However, such exhaustive patch-level processing is computationally expensive, severely limiting the efficiency and scalability of WSI analysis. To address this challenge, we propose PathCTM (a Pathology-oriented Continuous Thought Model) that enables token-efficient scale-space continuous reasoning for gigapixel WSIs. PathCTM formulates diagnostic inference as a dynamic sequential information pursuit. It progressively transitions from low-magnification global to high-magnification local inspection, and adaptively terminates inference when sufficient evidence is gathered to effectively bound decision uncertainty. Specifically, it uses conditional computation for dynamic scale switching with attention-guided region pruning, coupled with confidence-aware early stopping. Extensive experiments demonstrate that, compared with standard MIL-based methods, PathCTM reduces the number of required image patches by 95.95% and shortens inference time by approximately 95.62%, while maintaining AUC without degradation. Code is available at https://github.com/JSGe-AI/PathCTM.
Authors: Bangrui Xu, Ziyang Miao, Xuanhe Zhou, Yiming Lin, Zirui Tang, Xiaomeng Zhao, Fan Wu, Cheng Tan, Fan Wu, Bin Wang, Conghui He
Abstract: VLM-based OCR models have become the de facto choice for document parsing, as they can accurately extract page-level elements (e.g., paragraphs within individual pages) together with their bounding boxes and textual content. However, downstream applications such as RAG require coherent document-level information, whereas these models often break cross-page continuity and fail to recover disrupted structures, such as paragraphs and tables truncated by page boundaries. Such relationships are not confined to a single page; instead, they require joint analysis of titles, paragraphs, tables, and images spanning multiple pages. A natural solution is therefore to reuse existing OCR outputs and reconstruct document-level logical structures through post-processing. To this end, we propose MinerU-Popo, a lightweight and universal framework for POst-Processing OCR outputs, which converts page-level results from diverse parsers into coherent document-level structures. MinerU-Popo decomposes the problem into four focused subtasks: text truncation recovery, table truncation recovery, title hierarchy reconstruction, and image-text association. To address these effectively, we build a task-oriented data engine with task-specific input filtering, and use the generated data (30K) to fine-tune a lightweight post-processing model (Qwen3-VL-4B). To support long documents, we introduce dynamic chunking with overlap-based synchronization, which aligns chunk-level outputs from the fine-tuned model and preserves global consistency. Finally, we assemble the aligned outputs into a tree-structured document representation, further enriched with node chunking and summaries for downstream retrieval and analysis. Empirical results show MinerU-Popo improves title-hierarchy TEDS by at least 20% across all five tested OCR models, improves RAG accuracy and reduces per-query latency.
Authors: Yuhang Zhang, Junxiang Qiu, Huixia Ben, Zhenhua Tang, Lin Liu, Shuo Wang, Yanbin Hao
Abstract: Training-free feature caching accelerates diffusion transformer (DiT) inference by reusing or forecasting intermediate features. However, fixed schedules make compute predictable but ignore prompt- and timestep-dependent risk, whereas hand-tuned error thresholds adapt locally but leave realized compute difficult to control. To address these limitations, we present Budgeted Cache Refresh (BeCARE), a training-free framework that, given a user-specified cap on full computations, adaptively determines when to refresh the cache during accelerated inference. Specifically, we first derive an error-amplification profile from the sampler's noise schedule to characterize the varying impact of approximation errors across denoising timesteps. We then combine this profile with prompt-specific extrapolation residuals and cache age to form an accumulated risk score, triggering a refresh when continued caching becomes harmful. Meanwhile, we use the same profile to construct an analytic spending reference and adjust the refresh threshold through feedback, thereby distributing the limited refresh budget over the sampling trajectory. Finally, structural safeguards (a fixed warmup, a budget-derived late-stage reserve, and a maximum cache-reuse length) prevent unreliable extrapolation and premature budget exhaustion. These designs enable prompt-adaptive refresh placement and allow a single calibration to transfer across budget tiers without per-tier tuning. Experiments on FLUX.1-dev with 200 prompts and measured FLOPs show that our method consistently outperforms representative training-free caching baselines across speedups from 3x to 6x. At the main operating point (3.3x acceleration), it improves paired PSNR by 2.7 dB over the strongest baseline using no more compute. On SD3.5 Large, it also substantially improves the same Taylor cache engine over fixed-interval scheduling at matched compute.
Authors: Caio Petrucci, Leo Sampaio Ferraz Ribeiro, Sandra Avila
Abstract: Age estimation from facial images typically relies on training data that includes images of minors, a practice that raises ethical, legal, and privacy concerns and that child-data governance frameworks explicitly advise against. While the task remains relevant (e.g., for detecting child sexual abuse imagery), we advocate against using data from minors entirely and quantify what the exclusion costs in accuracy. We formalize age estimation without children's training data as a generalized zero-shot learning (GZSL) problem: age intervals present during training are seen classes and withheld intervals are unseen, with models evaluated jointly on both. The generalized setting, rather than conventional zero-shot evaluation on unseen classes alone, is the appropriate one here because a deployed estimator must operate across the entire lifespan, not only on the interval withheld from it. Revisiting six widely used datasets, we introduce standardized splits with strict age-group separation. For datasets with identity annotations, subject-age-exclusive splits prevent identity leakage across the seen/unseen boundary. Evaluating nine state-of-the-art age estimation methods under this protocol reveals that all of them fail to generalize to unseen age groups, suffering substantial degradation --- on average 46.4%, and up to 52.8% --- relative to the supervised baseline. Moreover, models do not simply degrade: they systematically anchor predictions for unseen ages to nearby seen classes, a manifestation of the well-known seen-class bias in generalized zero-shot learning.
Authors: Mojgan Forootan, Mahziar Setayeshfar, Ali Darvishi, Mohammad Tashakoripour, Hamidreza Bolhasani
Abstract: Gastric intestinal metaplasia (GIM) is a precursor lesion to gastric dysplasia and adenocarcinoma whose early detection is crucial for intervening in the carcinogenesis cascade. Artificial intelligence (AI) holds considerable promise for real-time endoscopic detection and characterization of GIM. However, development of reliable AI models has been constrained by the absence of publicly available, histopathologically validated datasets that combine detailed endoscopic annotations, histological subtype (complete and incomplete), standardized grading systems, and normal mucosal patterns. GIM-ENDO was designed to fill this gap. The dataset comprises demographic data, endoscopic findings, histopathological results, and H. pylori status acquired using the Olympus EVIS X1 system with white-light endoscopy (WLE) and image-enhanced endoscopy (IEE), including narrow-band imaging (NBI) and magnifying NBI (M-NBI), along with images and video clips from 24 patients (22 GIM-positive, 2 normal controls). Annotations cover six primary IEE endoscopic signs -- light blue crest (LBC), marginal turbid band (MTB), white opaque substance (WOS), TV pattern (Fusion), atrophy, and map-like erythema (MLE) -- plus two additional endoscopic findings (AHP and GA) recorded where present. GIM subtypes (complete and incomplete) are annotated for all GIM-positive cases; OLGA and OLGIM staging are provided where complete histological sampling was available. The dataset is publicly accessible at https://doi.org/10.5281/zenodo.20707267. For the latest updates and further information regarding this dataset, readers are referred to the DataBioX website: https://databiox.com A short version of this work has been submitted to MICCAI 2026 Open Data Track.
URLs: https://doi.org/10.5281/zenodo.20707267., https://databiox.com
Authors: Kunwar Maheep Singh, Christian Theobalt, Rishabh Dabral
Abstract: We present Generative Relightable Avatars (GRA), a person-specific method for photorealistic free-view rendering and environment-map relighting of full-body humans. We postulate that modeling fine-grained appearance details is inherently a one-to-many problem that can benefit from a generative formulation. In contrast to fully regressive relightable avatar methods, GRA follows a hybrid approach that combines controllable, physics-grounded relighting with probabilistic refinement. Starting from a tracked animated mesh, we optimize material parameters in UV-space and render a coarse relit appearance under a target HDR environment map. Next, we refine the textures with a feed-forward model to capture pose-dependent texture dynamics and illumination effects beyond simplified reflectance assumptions. Finally, a fine-tuned video-to-video diffusion model transforms the physically grounded renderings into temporally coherent, high-detail videos while preserving 3D control, with an error-recycling strategy for generating long videos. Experimental evaluations demonstrate our method's improved perceptual quality over prior relightable avatar baselines. Project Page: https://vcai.mpi-inf.mpg.de/projects/GRA/
Authors: Hojun Choi, Seulbin Hwang, Daejung Kim, Kisung Kim, Hyunjung Shim, Jinhan Lee
Abstract: Bird's-eye view (BEV) perception fuses multi-camera images into a unified top-down representation for autonomous driving. Despite recent progress, state-of-the-art methods remain confined to closed-set scenarios, making them vulnerable to unpredictable real-world environments. In this work, we introduce open-vocabulary BEV segmentation (OVBS), which leverages vision-language models (VLMs) to recognize categories beyond the training set while maintaining precise BEV perception and real-time efficiency. A key challenge in OVBS lies in the 3D geometric inconsistency inherent in the ill-posed lifting of 2D VLM semantics into BEV. To address this, we propose OVBEVSeg, a geometry-aware OVBS framework that enhances efficient Gaussian splatting (GS)-based unprojection by leveraging robust 3D geometric constraints across three progressive stages: (1) 2D-to-BEV pseudo-labeling via reliable 3D projection for OV generalization; (2) joint 2D-BEV per-scene optimization with BEV structural constraints for 3D geometric consistency; and (3) 3D geometric distillation for online efficiency. On the nuScenes dataset, OVBEVSeg achieves state-of-the-art performance, outperforming closed-set methods by 15.3 mIoU on unseen categories. Remarkably, even with no novel-class ground-truth labels, it remains competitive with self- and semi-supervised baselines trained with up to 40% of ground-truth annotations. Furthermore, it achieves 2.5x faster inference with only 0.22x the memory consumption of projection-based methods. Project page: https://hchoi256.github.io/projects/ovbevseg/.
Authors: Makoto Yamada
Abstract: Recently, Joint Embedding Predictive Architectures (JEPAs) have attracted significant attention in the computer vision and machine learning communities as a promising framework for self-supervised representation learning. Unlike masked autoencoders that reconstruct pixels, JEPA models learn representations by predicting latent embeddings of masked regions. Existing JEPA-based methods, such as I-JEPA and V-JEPA, typically employ a single encoder in the student network. In contrast, using Siamese encoders for student network is more naturally aligned with brain-inspired representation learning frameworks, yet their role in JEPA models remains largely unexplored. In this paper, we investigate the effect of Siamese student encoders in JEPA-based representation learning. To this end, we propose SiamJEPA, masked Siamese student encoders equipped with an exponential moving average (EMA) teacher network. SiamJEPA can also be viewed as a JEPA formulation of the brain-inspired representation learning model PhiNet. Through extensive experiments on ImageNet linear probing, we demonstrate that Siamese encoders act as an effective regularizer for the JEPA objective, improving representation separability and accelerating learning during the early stages of training. Furthermore, SiamJEPA consistently outperforms comparable single-encoder JEPA variants under limited training budgets and achieves higher linear probing accuracy than Masked Autoencoders (MAE) which requires longer training. Our findings reveal that Siamese student encoders are not merely an architectural choice but constitute an important inductive bias for predictive representation learning. These results provide new insights into the design of JEPA-based models and suggest that incorporating Siamese student architectures offers a simple yet effective approach for improving self-supervised representation learning.
Authors: Bingcai Wei, Yuning Cui, Mingyu Liu, Jinni Geng, Ling Li, Benwang Chen, Ziwei Li, Alois Knoll
Abstract: Real-world paired image dehazing remains challenging because haze degradation is spatially non-uniform, illumination-dependent, and physically ambiguous even when haze-free references are available. Existing end-to-end restoration networks usually learn a deterministic mapping from a hazy observation to a clean target, while degradation-sensitive feature responses, reverse haze-formation consistency, and cross-domain negative structure remain insufficiently exploited. In this paper, we propose Backbone-Agnostic Stochastic Perturbation Learning (BSPL), a plug-and-play framework for end-to-end real-world image dehazing. BSPL first introduces a Learnable Stochastic Perturbation Modulator (LSPM), which learns input-conditioned channel-wise and spatial-wise perturbation distributions and converts the resulting feature-response discrepancies into adaptive modulation weights. It then develops a Prior-informed Perturbation-guided Reconstruction Module (PPRM), which reuses the learned bottleneck perturbations together with transmission and atmospheric-light priors to reconstruct the hazy observation from the restored result and enforce degradation consistency. Furthermore, we propose a Dual-space Domain-diversified Distribution-aware Contrastive Loss ($D^3$CL) to regularize both clean restoration and hazy reconstruction spaces with real-world and synthetic negatives. Experiments on five real-world paired benchmarks show that BSPL consistently improves multiple representative backbones with only marginal additional inference overhead.
Authors: Zhanping Song, Shipeng Liu, Liang Zhao, Dengfeng Chen
Abstract: Cross-domain infrastructure crack segmentation is challenged by variations in materials, imaging conditions, crack morphology, and background interference. Although text-promptable foundation models reduce the need for task-specific training, SAM3's native proposal interface may not fully expose the crack evidence already encoded by the frozen model. We identify and quantify an output-interface mismatch in which the internal language-conditioned response preserves ground-truth-aligned crack evidence that is absent from the native proposal output. Based on this finding, we propose \textbf{S}emantic-\textbf{E}dge \textbf{R}esponse \textbf{D}ecoding (\textbf{SERD}), a training-free method that replaces proposal-based readout with direct response decoding. SERD normalizes the dense response, calibrates it using a fixed Sobel prior, and applies a single threshold to produce the crack mask. SAM3 remains frozen without any target-domain annotations. Using one source domain solely for threshold selection, SERD achieves an average Crack IoU of 58.00\% on five unseen datasets, compared with 54.33\% for native SAM3. Across six rotated source-domain settings, it obtains a mean target-domain IoU of 60.23\% and a Boundary F1 of 67.18\%, exceeding SAM3 by 3.27 and 2.70 percentage points, respectively. The results show that task-aligned internal-response decoding provides a simple and transferable solution for cross-domain infrastructure crack segmentation. \textit{Code is available at: \href{https://github.com/xauat-liushipeng/SERD}{GitHub}.}
Authors: Shuwei Huang, Tianyao Luo, Jicheng Liu, Pan Zhou
Abstract: Image super-resolution (ISR) has witnessed remarkable progress with diffusion models and flow matching. The dominant text-to-image (T2I) based approaches leverage large-scale foundation models as generative priors, achieving impressive perceptual quality but at the cost of massive model sizes and prohibitive training expenses. Recent flow-matching-based vision-only approaches have made significant strides; however, they adopt standard flow formulations that transport from a pure Gaussian prior to the data distribution, discarding the rich structural information already present in the low-quality (LQ) input. Furthermore, existing single-step acceleration techniques often forfeit the model's multi-step inference capability. In this paper, we propose Residual Flow Matching for Image Super-Resolution (RFMSR), a vision-only framework that centers the source distribution at the LQ latent, reducing transport distance and preserving structural priors throughout the flow trajectory. We further introduce a two-phase training strategy: Phase I pretrains the velocity field via conditional flow matching, while Phase II applies end-to-end supervision to the single-step prediction while retaining the velocity loss across all timesteps, achieving high-quality single-step generation without sacrificing multi-step refinement. Extensive experiments demonstrate that RFMSR achieves comparable or even superior perceptual quality compared to state-of-the-art (SOTA) methods. The source code is available at https://github.com/Faze-Hsw/RFMSR.
Authors: Zilu Zhou, Dongliang Chang, Junhan Chen, Zhanyu Ma
Abstract: Many fine-grained recognition tasks contain hierarchical labels such as order, family and species. Although this supervision should be beneficial, jointly optimising all levels often leads to unstable training because coarse and fine classifiers impose inconsistent gradients on the shared backbone. This hierarchical gradient conflict prevents the model from learning a coherent coarse-to-fine representation. In this paper, we propose FlexiGrad, a simple and parameter-free method that regulates gradient interactions during backpropagation. FlexiGrad removes only the harmful conflicting component when tasks disagree and reinforces the shared direction when they partially agree through a smooth hierarchy-aware weighting function. This produces stable optimisation and preserves both global structure and fine-grained discriminative cues. FlexiGrad integrates into existing architectures without modification while improves multi-granularity accuracy on CUB-200-2011, FGVC-Aircraft and Stanford Cars. The code will be available at PRIS-CV/FlexiGrad.
Authors: Maohua Li, Qirui Li, Yanke Zhou, Yiduo Li, Zhaosheng Chi, Chao Xu, Cuifeng Shen, Yixuan Xu, Hanlin Tang, Kan Liu, Tao Lan, Lin Qu, Shao-Qun Zhang
Abstract: Modern text-to-image diffusion transformers (DiTs) generate images through joint attention, in which text and image tokens interact directly within a single sequence. In large-scale DiTs, the conditioning input contains not only the user prompt but also chat-template tokens introduced by LLM-based text encoders. Yet how these tokens participate in the denoising computation remains poorly understood. To probe this, we introduce a causal interpretability framework. Using it to separate prompt-content tokens from chat-template tokens, we find that the template tokens carry little prompt-specific information at the encoder output. Yet surprisingly, they emerge as dominant image-to-text attention sinks and causally maintain object identity inside the DiT, acting as implicit semantic registers. We show that they acquire this identity indirectly. Rather than reading the prompt tokens, they draw the identity from the image latents into which the prompt semantics have already been injected at the very first layer. We further reveal a division of labor across heads and depth in DiTs, where distinct heads route semantics or render visual structure, and identity is committed in early blocks, carried by middle blocks, and refined in late ones. As a practical payoff, this analysis yields a training-free pruning rule that removes the causally inert prompt-reading heads and cuts $20\%$ of joint-attention FLOPs at a $1.4$-point cost in GenEval accuracy. Overall, our work not only reveals that the tokens encoding semantics at the input need not be those that maintain them during generation, but also provides a causal view of internal mechanisms in diffusion transformers.
Authors: Dankai Liao, Tianyi Zhang, Yufeng Wu, Xinyue Zhang, Qiaochu Xue, Zeyu Liu, Dachun Zhao, Linghan Cai, Yueming Jin
Abstract: Whole-slide image (WSI) diagnosis requires identifying diagnostically relevant regions, examining them across magnifications, and integrating multi-scale evidence. However, most existing pathology benchmarks evaluate models on pre-cropped patches or pre-extracted slide features, leaving their ability to acquire evidence directly from gigapixel WSIs largely untested. We introduce PathAgentBench, a benchmark for evaluating evidence-seeking vision-language models (VLMs) across four complementary capabilities: image-to-text matching for evidence interpretation, text-to-image retrieval for evidence verification, diagnostic-region localization for evidence acquisition, and multi-scale reasoning for evidence integration. The benchmark is organized as a diagnostic tree that links nested regions across magnifications with scale-specific findings and path-level diagnoses. It contains 1,822 TCGA WSIs and 17,135 diagnostic paths annotated by ten board-certified pathologists. An additional private cohort of 190 breast cancer WSIs with detailed annotations is used to evaluate autonomous whole-slide exploration. We evaluate 20 general-purpose, medical, and pathology-specialized models. Leading open-weight models achieve over 93% accuracy in multi-scale reasoning and over 50% accuracy in both cross-modal matching tasks. In contrast, diagnostic-region localization remains challenging: the best text-guided mean intersection-over-union is below 0.09, underperforming a simple center-based heuristic. During autonomous exploration, the unconditional hit rate decreases from 0.522 at low magnification to 0.185 at intermediate magnification and 0.020 at high magnification. These results reveal a pronounced gap between reasoning over curated evidence and acquiring that evidence directly from WSIs. PathAgentBench provides a unified framework for measuring and improving evidence-seeking pathology models.
Authors: Ziyao Huang, Shunkai Li, Juan Cao, Chenyu Li, Youliang Zhang, Zixiang Zhou, Cong Wang, Yuan Zhou, Qinglin Lu, Fan Tang
Abstract: Recent advances in video diffusion models have spurred interest in human-object interaction (HOI) video generation, which demands fine-grained control over interaction logic beyond single-subject animation. However, existing HOI methods rely heavily on explicit motion control, limiting scalability and generalization across diverse objects and interactions. In this study, we propose AgentHOI, a text-driven HOI video generation following a thinking-before-generation framework that bridges the gap between high-level textual intent and physical execution through multi-agent reasoning over perception, interaction, and motion planning. Building upon the generated interaction plans, we further strengthen text-driven motion understanding. We introduce an implicit text-motion alignment strategy that distills text-to-motion priors into the video diffusion model, enabling robust HOI synthesis without explicit motion inputs at inference. Experiments show that AgentHOI significantly improves interaction naturalness, object appearance preservation, and adherence to complex textual instructions across challenging object-centric scenarios such as wearing and riding. The code is available at https://github.com/bone-11/agenthoi.
Authors: Keyu Li, Jin Gao, Jialing Zhang, Dequan Wang
Abstract: Concept unlearning is increasingly used to limit the reproduction of protected or unsafe visual concepts in text-to-image models. Existing evaluations, however, mostly study targets that dominate the whole image, such as styles, broad object categories, or portrait-like identities, leaving company logos comparatively underexamined. Logos create a different failure mode: a small localized mark can carry the entire protected concept, must be visually precise to remain recognizable, and can be triggered implicitly by products, storefronts, packaging, or advertisements even when the word ``logo'' is absent. We introduce LU-500, a logo-unlearning benchmark built from Fortune Global 500 companies to study this localized and semantically entangled setting. LU-500 contains nearly 10,000 curated text-query and logo-image pairs, with an explicit track (LUex-500) and an implicit contextual track (LUim-500). To avoid reducing the task to a binary detector score, we define a multi-grained protocol that evaluates both local logo removal and global image preservation in pixel and latent spaces. Experiments on representative inference-time methods, including NP, SLD, and SEGA, and compatible fine-tuning-based methods such as ESD and Forget-Me-Not, show that the evaluated methods struggle to remove logo evidence without changing non-target content. We further analyze ProLU, a prompt-space multi-agent baseline: it improves local erasure by removing logo-inducing semantics, but also illustrates why prompt filtering is not a substitute for weight-level disentanglement. Correlation analyses over logo area, location, and structural complexity suggest that future logo unlearning may need spatially aware controls, such as SSIM-guided constraints, rather than purely global concept suppression.
Authors: Bingnan Li, Haozhe Wang, Haozhong Xiong, Fangtai Wu, Jinpeng Yu, Yang Shi, Jiaming Liu, Ruihua Huang
Abstract: On-policy distillation (OPD) adapts diffusion models by querying a teacher along trajectories generated by the current student, but how it should behave under classifier-free guidance (CFG), a default component of modern diffusion systems, remains poorly understood. Existing OPD methods naturally extend velocity matching to the CFG-composed prediction, directly matching teacher and student guided velocities. We show that this objective is under-identified at the branch level: positive- and negative-branch errors can compensate in the guided prediction. Through two contrasting cases, we find that naive matching remains effective under shared negative conditioning, where both branch errors decrease jointly. When the model's native CFG schema retains privileged information in the teacher's negative branch that is unavailable to the student, however, this joint reduction breaks down and the composed objective induces antagonistic branch-error dynamics, reducing the positive-branch error while increasing the negative-branch error. We term this failure mode Negative Branch Asymmetry (NBA). To address NBA, we introduce Positive--Direction Matching (PDM), a branch-aware OPD objective that separately constrains the positive prediction and the CFG conditional direction. We apply PDM to dense-to-sparse video control, where naive guided matching is highly sensitive to inference guidance scales, while branch-aware supervision enables more robust and effective knowledge transfer.
Authors: Yuyun Chen, Tianao Li, TianQuan Feng, Cen Chen, Huiping Zhuang, Hao Peng, Ziqian Zeng
Abstract: Reliable visual safety understanding in real-world scenarios demands more than just object recognition; it requires causal reasoning under epistemic uncertainty. While Large Vision-Language Models (LVLMs) demonstrate impressive semantic alignment on standard benchmarks, they often struggle to distinguish between superficial correlation and genuine forensic logic when grounded in the dynamic, partially observable nature of first-person experiences. Existing evaluations, dominated by third-person surveillance footage and binary classification metrics, fail to expose this cognitive gap. To address this, we introduce EgoSafe-Bench, a benchmark specifically designed to probe forensic reasoning in egocentric safety scenarios. It comprises 12,000 unique evaluation samples, generated by pairing each of the 3,000 video clips with a QA chain governed by our proposed Hierarchical Reasoning Evaluation (HRE) protocol. Unlike standard benchmarks, HRE mandates a rigorous reasoning trajectory from initial feature anchoring to blind-spot deduction and intent inference, thereby enforcing logical consistency and penalizing shortcut-based predictions. Extensive evaluations of state-of-the-art LVLMs (e.g., Qwen3-VL, Gemini, VideoLLaMA 3) reveal a significant perception-reasoning decoupling: models often achieve high descriptive scores but exhibit notable fragility in causal reasoning and logical closure. Our work provides both a challenging dataset and a systematic evaluation framework to foster the development of logically robust video understanding systems.
Authors: Ignacio M. De la Jara, Cristian Rodriguez-Opazo, Hamed Damirchi, Stephen Gould, Damith Ranasinghe
Abstract: Vision models do not form a representation at once; each block revises it. We ask whether the resulting computation path contains evidence that the final representation discards, and whether that evidence improves OOD detection and image classification on clean and shifted data. Unlike approaches that treat intermediate layers as separate snapshots, we retain sample identity across depth and study the transformations connecting successive states. We separate class-coherent transport from input-specific innovation, and coordinate movement from relational reorganization. Across supervised, self-supervised, vision--language, hierarchical, and convolutional encoders, these paths show strong sample-specific continuity and architecture-specific depth profiles that recur across datasets. They are also practically useful. An ID-only transition-surprise score complements strong final-state detectors, reducing FPR95 in 131/152 non-saturated comparisons on a balanced OpenOOD grid; gains are largest for visually disruptive and semantically far shifts, and remain positive on near-OOD for most detectors. Frozen update probes improve 71/72 clean model--dataset cases, while shifted-data gains vary with architecture and corruption type. Computation paths therefore provide a broadly useful reliability signal whose value is determined jointly by model organization and the shift encountered.
Authors: Paul Julius K\"uhn, Saptarshi Neil Sinha, Tiago Kleist, Richard Hoffmann, Arjan kuijper, Michael Weinmann
Abstract: While automated defect detection such as the detection of surface scratched is an important aspect in industrial quality control, the scarcity of annotated defect data make this task challenging. This paper presents a procedural rendering pipeline that generates large-scale annotated synthetic training data using BlenderProc, with configurable material appearance, camera modes, and domain randomization, producing automatic COCO-format annotations. To show the potential of our approach, we evaluate four training strategies, namely synthetic-only, real-only, mixed, and fine-tuning from synthetic weights, across two objects with different material properties and three lightweight edge-deployable detectors, YOLOX, YOLO26, and LW-DETR. Our evaluation show that fine-tuning from synthetic weights consistently outperforms real-only training, and that mixed training effectively recovers performance under scarce real-data conditions, with findings validated across both convolutional and transformer-based architectures. The proposed approach enables scalable defect detection without the burden of large real annotated datasets, making it practical for on-device industrial inspection. The pipeline scripts, 3D model, and both synthetic and real annotated scratch datasets for a glossy toy Ferrari car will be made available through the project website upon acceptance.
Authors: Feixiang Liu, Qiang Qiu, Lanbo Sun, Nan Wei, Huawei Shen, Xueqi Cheng
Abstract: Closed yes/no spatial benchmarks can reward a correct answer even when the image adds little support beyond no-image contexts. Under a fixed forced-choice interface, Visual Credit Audit (VCA) separates two estimands: whether the benchmark image gives the model's declared decision more support than text-only and blank controls, and whether the model responds to relation-specific visual evidence. The first audit is training- and label-free and does not require an answer flip. Applying labels yields dependence-credited correctness (D-CC); on correct items, it equals same-control gold-aligned positive gain, while prediction alignment extends the audit to errors. Across four open MLLMs and two spatial benchmarks, 12.73-26.25% of decisions are correct yet uncredited. Matched same-split image permutation reduces D-CC by 21.25-47.80 points, with every paired 95% interval above zero. Fixed-pixel relation contrasts and a 3x3 evidence-source factorial show why null controls cannot identify relation response. Among controlled correct-but-uncredited agreement decisions, response to relation reversal spans 81.57-100.00%, while 32.11% pooled change answer. Independently audited outcomes on 108 geometry-compatible edits provide a bounded natural-image correspondence check. VCA thereby decomposes benchmark success into correctness, additional image support, and relation-consistent response.
Authors: Wenhao Huang, Haifan Gong, Huan Zhang, Yu Wang, Xiang Wan, Haofeng Li, Guanbin Li, Hong Shen
Abstract: CT-based bronchial tree analysis is essential for diagnosing lung and airway diseases, yet automatic bronchus classification remains challenging because bronchial topology varies substantially across individuals. We propose the Bronchus Classification Network (BCNet), a structure-guided framework that uses segment-level topological information from point clouds to improve voxel-level representation learning. BCNet contains two jointly trained branches: a Point-Voxel Graph Neural Network (PV-GNN) for segment classification and a Convolutional Neural Network (CNN) for voxel-wise labeling. The branches share a common convolutional backbone, allowing topology-aware supervision from the PV-GNN to enhance voxel-level features. During inference, only the CNN branch is required, so BCNet retains the computational efficiency of its CNN baseline. Experiments on BronAtlas demonstrate that BCNet outperforms state-of-the-art methods by more than 8.0% in F1-score for bronchus classification. We also introduce BronAtlas, an open-access benchmark for bronchial imaging analysis that contains high-quality voxel-wise annotations of anatomical and abnormal bronchial segments. BronAtlas provides a valuable resource for developing and evaluating advanced methods for bronchial tree analysis, disease diagnosis, and surgical planning.
Authors: Ziran Zhang, Yuhang Tang, Zhigang Wang, Yueting Chen, Bin Zhao
Abstract: Infrared imaging and turbulence strength measurements are in widespread demand in many fields. This paper introduces a Physical Prior Guided Cooperative Learning (P2GCL) framework to jointly enhance atmospheric turbulence strength estimation and infrared image restoration. P2GCL involves a cyclic collaboration between two models, i.e., a TMNet measures turbulence strength and outputs the refractive index structure constant (Cn2) as a physical prior, a TRNet conducts infrared image sequence restoration based on Cn2 and feeds the restored images back to the TMNet to boost the measurement accuracy. A novel Cn2-guided frequency loss function and a physical constraint loss are introduced to align the training process with physical theories. Experiments demonstrate P2GCL achieves the best performance for both turbulence strength estimation (improving Cn2 MAE by 0.0156, enhancing R2 by 0.1065) and image restoration (enhancing PSNR by 0.2775 dB), validating the significant impact of physical prior guided cooperative learning.
Authors: Jameson Merkow, Felix J. Dorfner, Xiyu Yang, Alexander Ersoy, Giridhar Dasegowda, Mannudeep Kalra, Matthew P. Lungren, Christopher P. Bridge, Ivan Tarapov
Abstract: The integration of artificial intelligence (AI) into medical imaging has advanced clinical diagnostics but poses challenges in managing model drift and ensuring long-term reliability. To address these challenges, we develop MMC+, an enhanced framework for scalable drift monitoring, building upon the CheXstray framework that introduced real-time drift detection for medical imaging AI models using multi-modal data concordance. This work extends the original framework's methodologies, providing a more scalable and adaptable solution for real-world healthcare settings and offers a reliable and cost-effective alternative to continuous performance monitoring addressing limitations of both continuous and periodic monitoring methods. MMC+ introduces critical improvements to the original framework, including more robust handling of diverse data streams, improved scalability with the integration of foundation models like MedImageInsight for high-dimensional image embeddings without site-specific training, and the introduction of uncertainty bounds to better capture drift in dynamic clinical environments. Validated with real-world data from Massachusetts General Hospital during the COVID-19 pandemic, MMC+ effectively detects significant data shifts and correlates them with model performance changes. While not directly predicting performance degradation, MMC+ serves as an early warning system, indicating when AI systems may deviate from acceptable performance bounds and enabling timely interventions. By emphasizing the importance of monitoring diverse data streams and evaluating data shifts alongside model performance, this work contributes to the broader adoption and integration of AI solutions in clinical settings.
Authors: Qiao Yan, Yuchen Yuan, Xiaowei Hu, Yihan Wang, Jiaqi Xu, Xiwen Wu, Jinpeng Li, Chi-Wing Fu, Pheng-Ann Heng
Abstract: The increasing use of vision-language models (VLMs) in healthcare applications presents great challenges related to hallucinations, in which the models may generate seemingly plausible results that are in fact incorrect. Such hallucinations can jeopardize clinical decision making, potentially harming the diagnosis and treatments. In this work, we propose MedHallTune, a large-scale benchmark designed specifically to evaluate and mitigate hallucinations in medical VLMs. Comprising over 100,000 images and 1,000,000 instruction pairs, MedHallTune includes both hallucination and non-hallucination samples, each with ground-truth annotations. We conduct a comprehensive evaluation of current medical and general VLMs using MedHallTune, assessing their performance across key metrics, including clinical accuracy, relevance, detail level, and risk level. The experimental results show that fine-tuning with MedHallTune successfully improves the ability of several existing models to manage hallucinations and boost their zero-shot performance on downstream visual-question-answering (VQA) tasks, making them more reliable for practical medical applications. Our work contributes to the development of more trustworthy VLMs. Codes and dataset will be available at \href{https://github.com/russellyq/MedHallTune}{MedHallTune}.
Authors: Xiaoxu Zhu, Dongchuan Ran, Yiming Ren, Baoxiang Li
Abstract: Finite Scalar Quantization (FSQ) offers simplified training but suffers from residual magnitude decay in multi-stage settings, where subsequent stages receive exponentially weaker signals. We propose Robust Residual Finite Scalar Quantization (RFSQ), addressing this fundamental limitation through two novel conditioning strategies: learnable scaling factors and invertible layer normalization. Our experiments across audio and image modalities demonstrate RFSQ's effectiveness and generalizability. In audio reconstruction at 24 bits/frame, RFSQ-LayerNorm achieves 3.646 DNSMOS, a 3.6% improvement over state-of-the-art RVQ (3.518). On ImageNet, RFSQ achieves 0.102 L1 loss and 0.100 perceptual loss, with LayerNorm providing 9.7% L1 improvement and 17.4% perceptual improvement over unconditioned variants. The LayerNorm strategy consistently outperforms alternatives by maintaining normalized input statistics across stages, effectively preventing exponential magnitude decay that limits naive residual approaches. RFSQ combines FSQ's simplicity with multi-stage quantization's representational power, establishing a new standard for neural compression across diverse modalities.
Authors: Haiyang Li, Yaxiong Wang, Shengeng Tang, Yuchen Zhang, Lianwei Wu, Lechao Cheng, Liu Liu, Chaofeng Dong, Zhun Zhong
Abstract: Detecting deceptive multimodal content on social media has become an increasingly important problem. Two major types of deception dominate: human-crafted misinformation (e.g., rumors and misleading posts) and AI-generated content produced by image synthesis models or vision-language models (VLMs). However, these two types are usually addressed as separate tasks. Consequently, existing models are often specialized for only one type of fake content. In real deployments, however, the fake-content type of an incoming multimodal post is typically unknown, which limits the practicality of such specialized systems. To study this setting, we build OmniFake, a benchmark with 98K samples that combines human-curated misinformation from existing resources with newly created AI-generated examples. To address this new task, we propose Unified Multimodal Fake Content Detection (UMFDet), a framework designed to handle both types of deception. UMFDet builds on a VLM backbone augmented with a Category-aware Mixture-of-Experts (CMoE) adapter to capture category-specific cues. We further introduce an Expert-wise Discriminative Regularization to enforce intra-expert compactness. In addition, cross-modal consistency alignment is proposed to improve the perceptual capability of experts for handling different deception types. Experiments show that UMFDet consistently outperforms competitive specialized baselines across both deception types.
Authors: Soujanya Hazra, Rupam Mukherjee, Rajkumar Daniel, Shirin Dasgupta, Subhamoy Mandal
Abstract: Early detection of oral cancer and potentially malignant diseases is a major challenge in low-resource settings due to the scarcity of annotated data. We provide a unified approach for oral lesion classification that incorporates deep learning, spectral analysis, and demographic data. A pathologist verified subset of oral cavity images was curated from a publicly available dataset. Oral cavity pictures were processed using a fine tuned ConvNeXtv2 network for deep embeddings before being translated into the hyperspectral domain using a reconstruction algorithm. Haemoglobin sensitive, textural, and spectral descriptors were obtained from the reconstructed hyperspectral cubes and combined with demographic data. Multiple machine learning models were evaluated using patient specific validation. Finally, an incremental heuristic meta learner (IHML) was developed that merged calibrated base classifiers via probabilistic feature stacking and uncertainty-aware abstraction of multimodal representations with patient level smoothing. By decoupling evidence extraction from decision fusion, IHML stabilizes predictions in heterogeneous, small sample medical datasets. On an unseen test set, our proposed model achieved a macro F1 of 66.23% and an overall accuracy of 64.56%. The findings demonstrate that RGB to hyperspectral reconstruction and ensemble meta learning improve diagnostic robustness in real world oral lesion screening.
Authors: Filippo Rinaldi, Aniello Panariello, Giacomo Salici, Angelo Porrello, Simone Calderara
Abstract: Adapting large pre-trained models to downstream tasks often produces task-specific parameter updates that are expensive to relearn for every model variant. While recent work has shown that such updates can be transferred between models with identical architectures, transferring them across models of different widths remains unexplored. In this work, we introduce Theseus, a training-free method for transporting task updates across heterogeneous-width models. Rather than matching parameters, we characterize a task update by the functional effect it induces on intermediate representations. We formalize task-vector transport as a functional matching problem on observed activations and show that, after aligning representation spaces via orthogonal Procrustes analysis, it admits a stable closed-form solution that preserves the geometry of the update. We evaluate Theseus on vision and language models across different widths, showing consistent improvements over baselines without additional training or backpropagation. Our results show that task updates can be meaningfully transferred across architectures when task identity is defined functionally rather than parametrically. Code is available at https://github.com/apanariello4/merge-and-rebase.
Authors: Yuming Yan, Kai Tang, Sihong Chen, Ke Xu, Dan Hu, Qun Yu, Pengfei Hu
Abstract: Current post-training methodologies for adapting Large Vision-Language Models (LVLMs) generally fall into two paradigms: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL). Despite their prevalence, both approaches suffer from inefficiencies when applied in isolation. SFT forces the model's generation along a single expert trajectory, often inducing catastrophic forgetting of general multimodal capabilities due to distributional shifts. Conversely, RL explores multiple generated trajectories but frequently encounters optimization collapse - a cold-start problem where an unaligned model fails to spontaneously sample any domain-valid trajectories in sparse-reward visual tasks. In this paper, we propose Supervised Group Relative Policy Optimization (S-GRPO), a unified post-training framework that integrates the guidance of imitation learning into the multi-trajectory exploration of preference optimization. Tailored for direct-generation visual tasks, S-GRPO introduces Conditional Ground-Truth Trajectory Injection (CGI). When a binary verifier detects a complete exploratory failure within a sampled group of trajectories, CGI injects the verified ground-truth trajectory into the candidate pool. By assigning a deterministic maximal reward to this injected anchor, S-GRPO enforces a positive signal within the group-relative advantage estimation. This mechanism reformulates the supervised learning objective as a high-advantage component of the policy gradient, compelling the model to dynamically balance between exploiting the expert trajectory and exploring novel visual concepts. Theoretical analysis and empirical results demonstrate that S-GRPO gracefully bridges the gap between SFT and RL, drastically accelerates convergence, and achieves superior domain adaptation while preserving the base model's general-purpose capabilities.
Authors: Xin Zhang, Yabo Chen, Yijie Fang, Wanying Qu, Haibin Huang, Chi Zhang, Feng Xu, Xuelong Li
Abstract: Recent generative video models achieve impressive visual quality but remain constrained by limited physical consistency and controllability. Existing video generation methods provide minimal physical control, and single-image-to-3D conversion approaches often suffer from object interpenetration. Furthermore, physics-based scene-level 3D generation methods exhibit spatial misalignment, stylized artifacts, and inconsistencies with the input data, restricting their use in realistic interactive video synthesis. We propose PhysOmni, a training-free framework that converts a single image into a physically consistent and controllable video through holistic scene-level 3D reconstruction. By rep?resenting the full scene geometry in a unified spatial coordinate system, PhysOmni resolves object penetration and alignment ambiguity. Unlike prior methods, this formulation enables accurate scene?level multi-object interactions and introduces richer, complex control types for advanced mechanics?based manipulation. By decoupling simulation from rendering, PhysOmni bypasses latency-heavy priors, achieving real-time physical interaction previews paired while preserving photorealistic visual fidelity. Experimental results demonstrate that PhysOmni substantially outperforms prior methods in physical fidelity, spatial coherence, and controllability. Project Page: https://physomni.github.io/
Authors: Pedro Figueiredo, Zixuan Li, Beibei Wang, Milo\v{s} Ha\v{s}an, Nima Khademi Kalantari
Abstract: We propose a neural formulation for estimating the appearance of complex luminaires. We focus on challenging luminaires with complex light transport (e.g., small emitters enclosed by multiple specular layers) that are difficult for (bidirectional) path tracing. To this end, we use light tracing to construct paths from emitters to the exit surfaces and formulate appearance estimation as a distribution learning problem. Specifically, we model the probability density function (pdf) of outgoing radiance on the exit surfaces using a large normalizing flow network, and recover the outgoing radiance as the product of the estimated pdf and flux. To enable efficient inference, we distill the learned appearance into a lightweight MLP that directly estimates radiance on the exit surfaces. We additionally train a sampling network for effective direct illumination computation from the luminaire, and a blending network to composite the luminaire into the scene. Our formulation makes it feasible to render challenging luminaires using low sample counts in arbitrary scenes.
Authors: Jo\~ao Pinelo, Jo\~ao Gon\c{c}alves, Arun Shukla, Adriana Santos-Ferreira
Abstract: The Internal Waves Service screens the Sentinel-1 Wave-mode archive for internal solitary waves, routing detections to experts whose adjudication time is the resource the effort exists to conserve. Because attention is the cost of error, precision leads. Its classifier was trained and reported at a one-to-one class balance, fixed before the operational rate could be known. That rate has since emerged at roughly one scene in twenty, and a balanced-test score badly overstates the precision a validator meets. A model that scores 0.794 balanced-test precision scores 0.192 in real operation: the gap is a systematic artefact of reporting at the wrong prior, invisible to the metric most work quotes. We show the mismatch to be an evaluation problem in the costume of a training one at a fixed recall, prior correction and calibration cannot move precision, and answer it with a prior-matched reporting method based on three numbers: balanced-test, operational-prior, and real post-deployment, whose contrast is the honest measure. A precision-first, leakage-controlled development cycle then improves the classifier lever by lever, each promoted only against a pre-registered margin; negative variety and the aggregation head lifting, capacity paying once then stopping, calibration inert, so the honest negatives are as much a result as the gains. Holding recall at a floor of 0.80 and certifying against a sealed, single-read lockbox, the promoted model reports 0.927 precision at the operational prior; an out-of-time check confirms discrimination transfers to unseen periods while a fixed operating point does not. Prior-matched reporting, begin balanced, then move to the prior as the stream reveals it, transfers to any operational Earth-observation service bootstrapping a rare-event detector under a prior it has yet to discover.
Authors: Shreshth Saini, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
Abstract: Test-time search lets small video diffusion models rival larger ones, but costs 2-10x more. All candidates are fully denoised, although most are discarded. Training-free caching makes each rollout 2-3x faster at near-lossless quality. Composition is safe only if lossy caching preserves verifier rankings. We present the first study of whether caching corrupts candidate ranking in video test-time search. On Wan2.1-T2V-1.3B with an adaptive caching wrapper (~2x per-candidate speedup), ImageReward scores seed-matched cached and full rollouts. Median per-prompt Spearman rank correlation is 0.905, with 72% top-1 agreement on the VBench suite. VBench-2.0 replicates this result on a harder suite. Recomputing the cached winner at full compute retains 90-94% of the full-search gain. Errors cluster among near-tied candidates, making corruption self-limiting. This finding leads to CachedSearch. It explores every candidate with aggressive caching, then re-generates only the winner at full compute. At N=8, it captures 94.7% of best-of-N's gain at 63% of the cost. Capture rises with width. At matched budget, it searches twice as wide for 38% more gain. The result holds from 1.3B-14B across six models and four families: Wan, LTX, CogVideoX, and Hunyuan. Wan2.1-14B matches the 1.3B model's fidelity. Mid-trajectory pruning multiplies the exploration saving to 3.11x at 88.6% capture. Ports to other model families require recalibrating a single parameter, showing that fidelity tracks architecture rather than parameter count. CachedSearch is training-free, verifier-agnostic, and orthogonal to the search algorithm, making it a plug-in multiplier for test-time scaling.
Authors: Arghya Sinha, Trishit Mukherjee, Kunal N. Chaudhury
Abstract: Pretrained deep denoisers can be used to solve a wide range of model-based image reconstruction tasks via Plug-and-Play (PnP) and Regularization-by-Denoising (RED) algorithms, without retraining per task. These denoisers are trained only for single-step denoising. Using them as Image Reconstruction (IR) regularizers in an iterative process can destabilize reconstruction. A common failure mode is the peak-and-collapse behaviour: metrics such as PSNR improve for early iterations and then abruptly degrade, making these algorithms unreliable in practice. We propose a data-driven stabilization framework that (i) formalizes this instability of any IR operator through a local quantity and (ii) prevents collapse by regularizing this quantity adaptively, requiring no retraining or modification of the given pretrained network. Our key idea is to control the potentially unstable IR operator with a contractive operator whose stable iterates act as an anchor and prevent collapse. We further introduce an efficient family of trainable contractive operators that serve as strong anchors while remaining lightweight. Extensive experiments across proximal algorithms, denoiser architectures, noise levels, and imaging tasks show consistent, collapse-free performance and improved reliability of PnP and RED reconstruction.
Authors: Abhishek Pillai, Samir Kumar Nayak, Yuan Chen
Abstract: Computer-use agents (CUAs) increasingly act through desktop GUIs to complete long-horizon tasks. Current benchmarks primarily measure end-task success or single-frame grounding. Neither isolates whether a model can reconstruct the causal, task-relevant transition produced by an action- crucial for rejecting stale observations, verifying progress, and recovering from failure. This is difficult because inference, remote input, app rendering, and screenshot capture are asynchronous: the next observation may be delayed, occluded, transient, or unrelated, then misread as progress and carried into subsequent planning. We introduce Desktop-Delta Bench (DDB), an offline step-level benchmark with 2,013 human-verified instances from novel, multi-app Linux trajectories across ~15 applications and 50 task domains. DDB trajectories targets 3 failure dimensions- state verification, source tracking, and context-aware control- through 2 complementary tasks: 463 3-frame temporal-ordering instances, including 105 with a cross-trajectory decoy, and 1,550 before-after pairs labeled from 5 actions + its payload. We evaluate 8 closed and open-source model families across 32 ordering and 16 single-action settings, observing consistent gaps. Ordering remains unsaturated: best non-decoy and decoy exact-match rates are 65.1% and 65.7%. Task context improves decoy identification by 6.9 percentage points but reduces non-decoy exact match by 2.2 points; error analysis reveals systematic copying of the presented A-B-C order. Single-action results show that inferring the action family is harder than locating it: click F1 is 0.96 vs, 0.76 for drag, while recognized drags are generally localized well. DDB, thus, complements end-to-end benchmarks by filling the missing diagnostic layer between GUI grounding and final task success, enabling targeted improvements to desktop CUA verification, reliability, and recovery.