VQA
-
CARVE: Cross-Slice Anisotropic Reallocation of Visual Evidence for Efficient 3D Medical Volume Understanding
2026-08-05 06:49:49Zhenyu Yi, Qiang Hu, Zhenhao Li, Jiaxuan Zhao, Yusong Sun, Lichi ZhangAbstract
Slice-based MLLMs leverage mature 2D encoders by representing 3D volumes as sequences of 2D slices. However, this slice-wise formulation produces thousands of visual tokens that burden the LLM backbone, many of which capture overlapping visual evidence across adjacent slices. To understand how effectively a growing visual token budget improves performance, we perform scaling analyses on two 3D medical VQA benchmarks and find diminishing returns: cost keeps rising while accuracy saturates, and improving in-plane resolution is more effective than adding slices at comparable budgets. The budget should therefore be allocated more selectively rather than simply enlarged, yet most token compression methods are designed for 2D images or videos, where redundancy arises from spatial layout or temporal motion rather than from near-duplicate content along the depth axis. We present CARVE, a training-free framework that compresses visual tokens prior to LLM inference and casts token reduction as budget-constrained 2.5D allocation. CARVE partitions the depth axis into coherent windows and allocates tokens non-uniformly according to normalized cross-slice evidence. Under a shared budget, CARVE builds spatial anchors on representative slices and retrieves locally varying evidence from the full volume, then merges remaining eligible tokens into nearby anchors within each window. Removing roughly 80% of the visual tokens on Hulu-Med-7B, CARVE leads all compression baselines on every AMOS-MM report-generation metric, with 6.2 points higher retention of full-token quality than the strongest baseline, and preserves 98.1% of full-token performance across three VQA benchmarks.
Abstract (translated)
URL
https://arxiv.org/abs/2608.04515
PDF
-
Q-CueGraph: Query-Conditioned Visual Evidence Graphs for Multimodal Reasoning
2026-08-05 05:03:35Pengcheng Pan, Xinfang ZhangAbstract
High-resolution pixels and crop or zoom tools give multimodal large language models the ability to inspect an image, but they do not provide a reliable task-conditioned policy for deciding where to inspect. Q-CueGraph makes this decision explicit. It maps a question and an image representation to budgeted, coordinate-level observations for a frozen reader. Text-rich images use a reusable OCR/layout graph; natural-image search instantiates query-conditioned visual nodes behind the same selection, composition, and budgeting interface. Optional utility refinement learns which candidate crops the frozen reader can use from training-answer correctness, without region-box supervision. With a frozen Qwen2.5-VL-7B reader, Q-CueGraph reaches 0.833 accuracy on V*Bench versus 0.696 for full-image inference from a 19% image-area budget, and reaches 92% of full-image ANLS on InfographicVQA from about half the image area. Across six benchmarks, explicit observation is most valuable when evidence is localizable, the question discriminates its location, and resolution limits full-image reading.
Abstract (translated)
URL
https://arxiv.org/abs/2608.04452
PDF
-
Video-DeepResearch: Towards the Next-Generation Multimodal Deepresearch Agent
2026-08-04 17:45:16Zhen Fang, Yu Zeng, Wenxuan Huang, Yiming Zhao, Shiting Huang, Tianfei Ren, Qi Lu, Qingnan Ren, Qisheng Su, Lionel Z. Wang, Qingyu Yin, Shuang Chen, Zehui Chen, Lin Chen, Zhenfei Yin, Yao Hu, Shaohui Lin, Wanli Ouyang, Shaosheng Cao, Feng ZhaoAbstract
We introduce Video-DeepResearch (Video-DR), extending multimodal agents from static images to continuous video streams, a setting that demands dense spatiotemporal grounding coupled with open-web exploration. Preliminary evaluations reveal two critical bottlenecks in current models: (1) modality bias, where agents bypass visual tools in favor of textual search, and (2) parametric knowledge leakage, where models rely on internal memory rather than genuine tool-augmented execution. To address these challenges, we propose Video-DR, featuring a decoupled perception-exploration pipeline with stage-wise tool unlocking that compels exhaustive cross-frame visual grounding prior to web retrieval. Our framework adopts a two-stage training recipe: supervised fine-tuning followed by Group Relative Policy Optimization (GRPO), enabling autonomous exploration that breaks the imitation-learning ceiling. Furthermore, we curate Video-DR-Bench, a human-AI collaborative benchmark comprising 200 complex, multi-hop VQA instances. Empirical results demonstrate that our Video-DeepResearch-35B-A3B establishes a new state-of-the-art of 64.0% average accuracy, surpassing proprietary Claude-4.5-Sonnet (59.0%) by 5.0 points and significantly outperforming GPT-5 (52.5%) and Gemini 2.5 Pro (57.5%). The 30B-A3B variant achieves 59.3%, competitive with Claude-4.5-Sonnet and demonstrating the effectiveness of our training paradigm even at compact scale. Code: this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03979
PDF
-
CARE-X: Towards Clinically Useful Radiology VLMs with Auxiliary Supervision, Reward-Aligned Learning, and Tool-Augmented Measurement
2026-08-04 16:23:39Mercy Prasanna Ranjit, Anirban Porya, Sathvik Joel, Niharika Vadlamudi, Nikhilesh Chowdary Eathamukkala, Prasanth V V, Abhyuday Kumara Swamy, Pranay Narhari Umredkar, Pradeep Narayan, Vivek Rajagopal, Tanuja GanuAbstract
A clinically useful chest X-ray system must go beyond fluent report generation: it should classify findings with tunable decision thresholds, localize them spatially, and derive the anatomical measurements upon which many diagnoses depend. Today's Vision-Language Models (VLMs) treat these as separate problems, if they address them at all, leaving a gap between what radiologists need and what generative models provide. We introduce CARE-X, a chest X-ray VLM that narrows this gap by unifying auxiliary discriminative supervision with reward-aligned generation. CARE-X augments its generative backbone with focal-loss classification and composite-loss grounding heads, co-trained alongside the language-modeling objective. This auxiliary supervision produces discriminative diagnostic predictions with tunable decision thresholds and precise spatial localization while also improving report quality, providing evidence that structured prediction and generation reinforce one another. Building on this foundation, Decoupled Clip and Dynamic Sampling Policy Optimization (DAPO) leverages task-specific reward signals for report generation, visual question answering (VQA), and spatial grounding, directly optimizing the clinical quality metrics that matter in practice. The result is state-of-the-art performance on the majority of metrics across four report-generation benchmarks, 94.0% VQA accuracy on ReXVQA (+6.0 pp over the next-best baseline), and generative spatial decoding that reaches near parity with dedicated detection heads. Separately, to address measurement-dependent diagnoses, we couple Qwen3-VL-4B-Instruct with native tool-calling capabilities for invoking deterministic measurement tools, while retaining full visual access to the image. This hybrid inference yields +43.6 pp average F1 over perception-only baselines across five measurement-dependent conditions.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03890
PDF
-
Geo-Embed: Towards Unified Multimodal Embeddings for Urban Understanding
2026-08-04 15:36:17Jiapeng Li, Yong Li, Junjie Zhou, Fan Zhang, Yu LiuAbstract
Geospatial and urban applications increasingly require models to compare heterogeneous evidence across street-view imagery, remote-sensing observations, text descriptions, region proposals, and temporal change cues. However, existing multimodal embedding models and benchmarks are still largely designed and evaluated around general-purpose image-text matching, leaving unclear whether unified embedding space can support heterogeneous geospatial tasks involving spatial relationships, fine-grained semantics, and temporal changes. To address this gap, we make three key contributions. First, we introduce GeoMEB, a large-scale multimodal embedding benchmark that standardizes 45 urban evaluation tasks across retrieval, visual question answering, change detection, classification, and visual grounding, together with training collections comprising 1.32M examples and 286K evaluation queries. Second, we present Geo-Embed, a unified embedding model that adapts a shared vision-language backbone to instruction-conditioned query-target matching over heterogeneous geospatial inputs, including single images, multiple images, text, regions, and masks. On GeoMEB, Geo-Embed achieves the strongest overall performance among representative multimodal embedders, with a 15.3% relative improvement over the strongest baseline. These results motivate future geospatial embedders that organize training and evaluation around explicit query-target relations, including semantic, cross-view, region-level, and temporal correspondence.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03826
PDF
-
Failure-Informed Image Self-Augmentation for Multimodal Large Language Model Self-Improvement
2026-08-04 14:28:42Chunyang Jiang, Pingping Zhang, Yuzhi Zhao, Wenao Ma, Zhijian Hou, Mengyang Wu, Yiyang Cai, Senkang Hu, Sitong Cheng, Chi-Min Chan, Wei Xue, Yike GuoAbstract
Multimodal large language models (MLLMs) have achieved remarkable performance across vision-language tasks, but their progress depends heavily on large-scale, high-quality multimodal data that are costly to annotate. Self-augmentation offers a promising alternative by enabling models to expand their own training data without external supervision. However, existing MLLM self-augmentation methods are largely text-centric, while image augmentation remains underexplored and typically relies on generic or handcrafted transformations that are weakly aligned with the model's actual incapability. We propose Failure-informed Image Self-Augmentation (\textbf{FISA}), a framework for MLLM self-improvement that constructs augmented images from the model's own failure cases. Our method generates visually challenging yet answer-preserving image complications, verifies their utility through self-examination, and applies dual fidelity filtering to avoid semantic distortion. Experiments on visual question answering benchmarks show that the proposed method consistently improves performance across both in-distribution and out-of-distribution settings. Further experiments validate the compatibility of FISA with existing textual self-augmentation approaches, the superior data efficiency of the synthesized samples over generic image augmentation baselines, and the practical effectiveness of the proposed filtering strategy.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03733
PDF
-
From Multi-Resolution Cells to Gigapixel Whole Slide Images Foundation Model for Computational Pathology
2026-08-04 11:51:16Basit Alawode, Moshira Ali Abdalla, Dwarikanath Mahapatra, Muhammad Muzammal Naseer, Sajid JavedAbstract
Vision Transformers (ViTs) and their hierarchical variants have achieved strong performance in Computational Pathology (CPath). However, most are pre-trained on single-resolution Whole Slide Images (WSIs), limiting their generalization across arbitrary resolutions. Gigapixel WSIs inherently contain diagnostic patterns at multiple scales, including cellular morphologies, tissue architectures, and global context, mirroring how expert pathologists examine WSIs. We introduce Multi-Resolution Pyramid Transformer (MRPT), a model that hierarchically aggregates multi-resolution information from cellular to tissue and WSI levels. MRPT employs a biologically meaningful Consecutive Cross-Resolution Attention (CCRA) mechanism to capture scale-independent interactions and enforces multi-resolution semantic consistency by aligning embeddings across resolutions, yielding robust and generalizable WSI representations. Pre-trained in a multi-resolution self-supervised manner on 624M patches, 2.4M regions, and 36K WSIs, MRPT learns rich coarse-to-fine histopathology features. Extensive experiments on 34 diverse datasets show that MRPT surpasses recent foundation models and Multimodal Large Language Models (MLLMs) in cancer subtype classification, tissue phenotyping, and Visual Question Answering (VQA) for WSI understanding.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03508
PDF
-
DocTrace: Towards Traceable Long Document VQA via Hierarchical Evidence Graph Reasoning
2026-08-04 08:04:59Le Xiang, Zhicheng Guan, Hong Chen, Xiaocong Lin, Zhenghua Lei, Teng Hu, Bolei He, Long ZengAbstract
Long Document Visual Question Answering (LongDocVQA) requires Multimodal Large Language Models (MLLMs) to locate, integrate, and reason over heterogeneous document elements distributed across multiple pages. Existing approaches, including end-to-end MLLMs, retrieval-augmented generation (RAG) pipelines, and document agents, often lack explicit mechanisms to represent and verify how grounded evidence is progressively composed during reasoning, limiting both answer accuracy and traceability. In this paper, we cast LongDocVQA as an explicit evidence graph reasoning problem rather than implicit answer prediction. To this end, we propose DocTrace, a hierarchical framework that progressively performs evidence localization, structured document parsing, and evidence graph reasoning to enable explicit evidence provenance. To effectively learn these capabilities, we develop a two-stage training framework: joint Supervised Fine-Tuning (SFT) first initializes evidence localization and graph reasoning abilities, followed by task-specific Group Relative Policy Optimization (GRPO) with dedicated rewards to further optimize these capabilities. Extensive experiments on MMLongBench-Doc, LongDocURL, and SlideVQA demonstrate that DocTrace consistently outperforms both existing open-source baselines and proprietary MLLMs. Compared with the Qwen3-VL-8B-Instruct backbone, DocTrace achieves absolute improvements of 14.4, 11.3, and 11.7 points on the three benchmarks, respectively. Beyond competitive performance, DocTrace constructs traceable evidence graphs with explicit node-level provenance, enabling transparent and verifiable reasoning for long document understanding.
Abstract (translated)
URL
https://arxiv.org/abs/2608.03292
PDF
-
Rectify Then Diffuse: Disentangling Concepts Before Denoising Trajectory Unfolds
2026-08-04 05:03:17Ning Zhu, An Chen, Mengfei Zhao, Juntao Xu, Jingze Liang, Boyuan Gu, Liang-Jian DengAbstract
Text-to-image diffusion models can generate individual concepts well, but they often omit or merge concepts incorrectly with multiple concepts. We trace these failures to an early coordination bottleneck: before denoising begins, prompt-conditioned attention may allocate different concepts to strongly overlapping spatial support, which can keep their attention coupled as denoising proceeds. This observation motivates treating compositional generation as a boundary-condition problem rather than repeatedly controlling the evolving trajectory. To this end, we propose Rectify-then-Diffuse (RTD), a training-free framework that rectifies the initial allocation once before standard denoising. Firstly, we propose Soft-Overlap Disentanglement (SOD), which converts normalized overlap between pilot concept maps into a differentiable and layout-agnostic separation objective. Secondly, we introduce Isotropic Gradient Rectification (IGR), which normalizes the SOD gradient and applies a bounded latent displacement with a consistent scale across prompts and initializations. Extensive experiments show that RTD achieves state-of-the-art compositional fidelity and robust gains. On the AE-Bench object pair subset, RTD improves BLIP-VQA by 45.8% and ImageReward by 19.6% over CO3 while running 2.3$\times$ faster. Code will be released at this https URL
Abstract (translated)
URL
https://arxiv.org/abs/2608.03135
PDF
-
Qwen-3D: A Generalist 3D Vision-Language Model for Spatial Understanding
2026-08-04 00:32:09Lucy Lin, Ayush Jain, Yifan Liu, Katerina FragkiadakiAbstract
Large Multimodal Models (LMMs) have achieved remarkable success on images and short videos, yet scaling them to long videos remains challenging due to frame-centric tokenization and limited context windows. 3D geometry provides a natural compression mechanism for visual streams: depth and camera pose enable observations from multiple views and time steps to be fused into a persistent, world-aligned representation. While recent 3D LMMs leverage geometry-aware representations to improve spatial reasoning, they continue to lag behind specialist 3D perception systems on grounding and segmentation tasks. We argue that a key limitation is geometry-aware decoding: existing methods communicate 3D predictions through language tokens, proposal selection, or lightweight grounding queries, creating a bottleneck between language reasoning and dense geometric prediction. Building on these insights, we introduce Qwen-3D, a geometry-aware LMM that compresses visual information within the Qwen backbone using multi-view geometric cues, enabling efficient long-horizon visual reasoning over static scenes. Qwen-3D augments visual tokens with 3D Rotary Positional Embeddings, allowing attention to operate directly in 3D scene space rather than across independent image frames and thereby facilitating scalable cross-view and temporal reasoning. To bridge language and geometry, Qwen-3D incorporates a query-based segmentation decoder that grounds language directly in the underlying 3D scene representation, unifying referential grounding, instance segmentation, and visual question answering across both images and videos. Across a diverse set of benchmarks, Qwen-3D surpasses existing 3D LMMs and outperforms several large proprietary 2D models. Notably, Qwen-3D achieves these improvements while maintaining strong performance on standard 2D vision-language benchmarks by jointly training on 2D and 3D data.
Abstract (translated)
URL
https://arxiv.org/abs/2608.02980
PDF
-
In-Context Collapse in Vision-Language Models and How to Mitigate it?
2026-08-03 19:44:45Mohammad RostamiAbstract
Many-shot in-context learning (ICL) lets vision-language models (VLMs) adapt from image--label demonstrations without weight updates, and is widely assumed to improve as more demonstrations are supplied. We show the opposite: as demonstrations accumulate, a subset of VLMs undergo an \emph{in-context collapse}, a sharp, sometimes catastrophic accuracy drop spanning synthetic classification, natural-image classification, and VQA benchmarks, in some models falling below chance while outputs remain well-formed. Across an open VLM panel ($0.5$B--$11$B) and a frontier model (Claude Sonnet 4.5), the collapse is graded. Two capabilities turn out to be dissociable: robustness to accumulating demonstrations and the ability to learn a novel rule in context, their combinations yield three reproducible regimes. A parameter-matched lesion-and-rescue causally localizes the collapse to the vision-language integration pathway: an adapter on the connector and early/mid layers restores genuine learning (remap accuracy $0.39!\rightarrow!0.91$ at 16 shots), while an equal-capacity adapter on the late readout does not. We propose \textsc{CircA}, whose core is a one-time integration vaccine: trained once on one synthetic task, it transfers collapse-resistance to unseen task families (chance$\rightarrow$$0.71$/$0.60$ on CIFAR/Fashion). The layers best for in-context integration are not the layers best for weight-based consolidation, the late readout achieves higher accuracy and less forgetting at fewer parameters. The collapse is an integration failure at the vision--language interface, correctable by a lightweight, transferable intervention.
Abstract (translated)
URL
https://arxiv.org/abs/2608.02830
PDF
-
A General-Purpose VLM Can Teach an Astronomy Foundation Model to Better Recognize Galaxy Morphology
2026-08-03 14:26:25Dichang Zhang, Jiaqi Deng, Yixuan Shao, Yuanpeng Liu, Jiali Cui, Zhiqiang Lao, Heather Yu, Liang Peng, Simon Birrer, Dimitris SamarasAbstract
Existing astronomy foundation models provide strong galaxy representations, but adapting them to new survey conditions and survey-specific morphology recognition tasks still requires substantial human supervision. We show that VLM-based VQA systems contain meaningful visual-semantic priors that can serve as weak supervision for downstream morphology classifiers and improve morphology classification under limited human-label budgets. We first introduce a survey-oriented VQA benchmark spanning two representative imaging regimes and evaluate state-of-the-art VLMs on galaxy morphology questions. The results show that these models capture useful morphology signals and informative uncertainty, but are not sufficiently reliable to replace human annotators. Motivated by this finding, we use a general-purpose VLM as a morphology teacher for Zoobot, an astronomy foundation model pretrained on large-scale Galaxy Zoo annotations. Across two survey domains and multiple annotation budgets, the VLM teacher consistently improves Zoobot's downstream morphology classification. These results demonstrate that a general-purpose VLM provides knowledge complementary to an astronomy foundation model and can teach it to better recognize galaxy morphology under limited human supervision. The resulting pipeline is designed for label-efficient adaptation to forthcoming large-scale surveys, including the Vera C. Rubin Observatory's Legacy Survey of Space and Time (LSST) and the Nancy Grace Roman Space Telescope. The benchmark and code are publicly available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.02300
PDF
-
HAFI-VLM: A Frequency Perspective for Diagnosing and Enhancing Visual Perception in Vision-Language Models
2026-08-03 12:15:33Jin Cui, Chuanchang Su, Jiayi Lu, Xinyue Long, Boran Zhao, Pengju RenAbstract
Vision-language models (VLMs) remain unreliable when predictions require fine-grained visual evidence. We identify a previously overlooked cause: spectral response rigidity. Despite substantial frequency variation across images and tasks, pretrained vision encoders exhibit persistent, encoder-specific layerwise spectral profiles that change only marginally under downstream fine-tuning. Since pretrained vision encoders only receive images, they cannot adapt spectral extraction to the evidence required by the current query. We therefore propose HAFI-VLM, which introduces a task-conditioned frequency pathway while preserving the pretrained semantic representation. Hierarchical Adaptive Frequency Injection (HAFI) retrieves complementary low-, mid-, and high-frequency evidence at multiple encoder depths using text-modulated, spatially aligned cross-attention. A Visual Enrichment Layer Adapter further recalibrates shallow LLM attention to effectively utilize the enriched visual tokens. Experiments on LLaVA-1.5 and Qwen2.5-VL demonstrate consistent improvements in general VQA, text-rich understanding, and hallucination robustness, outperforming representation-level enhancement methods and most resolution- or cropping-based approaches without additional high-resolution encoding. Mechanistic analyses show that HAFI restores task-dependent spectral allocation while retaining semantic attention, establishing frequency enrichment as a distinct and effective route for improving VLM perception.
Abstract (translated)
URL
https://arxiv.org/abs/2608.02124
PDF
-
Slot2Text: Object-Centric Visual Tokenization for Efficient and Spatially Traceable Surgical MLLMs
2026-08-02 20:09:45Guiqiu Liao, Matjaz Jogan, Daniel A. HashimotoAbstract
Multimodal large language models (MLLM) for surgical scene understanding typically inject hundreds of dense visual tokens into a language model, leading to costly inference and limited spatial traceability for generated answers. We present Slot2Text, a dual-mode surgical MLLM that replaces dense representations of visual input with a compact set of regions encoded as slot latents. Instead of relying on contrastive alignment of the visual encoder with language, Slot2Text groups self-supervised vision features into a few regions--slots that are consumed by the language model as area-labeled visual tokens. Slot2Text-Fast uses the slot prefix to answer surgical questions. Slot2Text-Reason also identifies and locates areas relevant for reasoning, linking language outputs to corresponding slot tokens, masks or regions. Experiments on multiple visual question answering and visual grounding benchmarks show that Slot2Text-Fast is competitive with state-of-the-art baseline at a much lower cost, reducing the average total token consumption by a 91.8\% and the visual prefix from 1,295 to 47 tokens (a 96.4\% reduction). Slot2Text-Reason trades additional tokens and latency for explicit area identities, locations, and traceable spatial evidence. These results establish compact slot latents as an efficient default visual interface for surgical MLLMs, with grounded reasoning invoked when greater spatial traceability is required.
Abstract (translated)
URL
https://arxiv.org/abs/2608.01473
PDF
-
LongChart VQA: A Comprehensive Benchmark for MLLMs with Complex Multi-Chart Reasoning
2026-08-02 15:48:04Ziyan Xiao, Yinghao Zhu, Wenting Zhang, Heaju Kim, Lequan YuAbstract
Multimodal large language models (MLLMs) are rapidly evolving with expanded context windows and stronger reasoning capabilities, enabling multi-chart understanding and multi-step inference. These abilities are increasingly important as MLLMs are adopted in complex agentic tasks. However, existing benchmarks largely emphasize single-chart perception, while simple chart-to-chart connections are insufficient to evaluate these capabilities. To capture multi-chart complexity while ensuring consistency and validity, we design a synthesis pipeline supported by latent graphs. Building on this pipeline, we introduce LongChart, a benchmark whose VQA sets contain an average of 6.5 images and 31.2 questions. We evaluate 10 state-of-the-art MLLMs and examine three factors that influence performance: reasoning patterns, auxiliary tools, and robustness to image perturbations. Our results show that MLLM accuracy decreases and varies substantially as computational complexity increases, highlighting directions for future research in multi-chart reasoning.
Abstract (translated)
URL
https://arxiv.org/abs/2608.01328
PDF
-
It's the Decoding Format, Not the Perturbation: Auditing Consistency-Based Selection for Vision-Language Test-Time Scaling
Abstract
Test-time scaling lifts large language model reasoning by sampling many candidate solutions and selecting among them, yet the same recipe transfers poorly to vision-language models (VLMs): recent work shows that simple majority voting beats selection methods built on the model's own self-verification, apparently because at the selection layer an image-grounded answer and a confident guess from the language prior look the same. A natural fix is to make the selection signal one that cannot be computed without the image. We study Perturbation Grounded Selection (Pgs), a label-free, training-free rule that scores each candidate by whether the model re-derives it under label-preserving perturbations of the input (cropping, background masking, mild photometric or geometric jitter); Pgs recovers majority voting when the perturbation set is empty. The decisive question is not whether Pgs beats chain-of-thought only majority voting, but whether the perturbation term adds anything once decoding format and budget are controlled. We therefore introduce a format-matched control (MatchedCtrl): the same short, no-CoT draws spent on the original image. Across TextVQA, MATH-Vision, MMMU, and ViLP, with a Qwen headline (three-seed means) and LLaVA-OneVision coverage in matched-budget selector tables, Pgs appears to beat plain majority voting by up to +31.8 points on TextVQA (Qwen), but MatchedCtrl tracks or exceeds Pgs within noise on every benchmark, including the vision-required ViLP; no Qwen category shows a significant gain over this control. The stability gap is real and image-dependent (up to +0.48), yet does not predict per-instance wins. The result is negative and diagnostic: perturbation consistency is at best a partial diagnostic of visual dependence and, on its own, not a usable selection signal once format is controlled; gains reported against CoT-only majority voting overstate such methods.
Abstract (translated)
URL
https://arxiv.org/abs/2608.01207
PDF
-
UniHEAR: Unified Heterogeneous-Source Attentive Retrieval for Knowledge-Based Visual Question Answering
2026-08-02 10:57:53Ganzhong Luo, Yang Ren, Hanyong Wang, Shuyu Zheng, Menglong YangAbstract
Knowledge-Based Visual Question Answering (KB-VQA) requires retrieving relevant entity knowledge from external sources to answer visually grounded questions. Existing retrieval-augmented systems suffer from two critical limitations. First, relying on a single retrieval modality creates a Single-Source Retrieval Bottleneck, missing ground-truth entities that are only accessible through complementary sources. Second, dual-tower pointwise rerankers suffer from Retrieval-Source-Blind Reranking, as they overlook retrieval origins and candidate-level retrieval priors, leading to redundant modality reliance. To address these challenges, we propose UniHEAR, a unified lightweight framework for heterogeneous-source entity retrieval and reranking. UniHEAR constructs a Coarse Retrieval Descriptor for each candidate entity, and introduces Retrieval-Guided Attentive Modality Gating to condition modality attention weights on this descriptor, further complemented by Entropy-Weighted Source Fusion of coarse retrieval priors. A hybrid training strategy combining contrastive learning with an auxiliary modality-preserving loss unifies entity-level and section-level retrieval within a single model. Extensive experiments on E-VQA and InfoSeek demonstrate that UniHEAR achieves state-of-the-art retrieval and VQA performance, improving Recall@1 by 6.7 and 1.2 points over the strongest baselines while maintaining a lightweight reranking architecture. Code and model are available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.01147
PDF
-
Location-Aware Fine-Grained Representation Learning for Medical Vision Foundation Models
2026-08-02 03:59:07Myeongkyun Kang, Yanting Yang, Xiaoxiao LiAbstract
Fine-grained visual representations are essential for medical image analysis, particularly when diagnostically relevant evidence is subtle and spatially localized. Modern transformer-based medical vision encoders must therefore learn patch-level representations that are both clinically meaningful and spatially consistent. Without these properties, large vision-language models (LVLMs) operate on an ambiguous visual foundation, limiting their ability to generate clinically reliable and spatially grounded responses. However, existing training strategies for medical vision encoders rarely achieve both objectives. Image-text alignment provides clinically meaningful supervision primarily at the image level, leaving the spatial localization of diagnostic evidence weakly constrained. In contrast, self-supervised learning promotes spatial consistency but lacks the semantic supervision needed to distinguish visually similar yet clinically distinct regions. To address this gap, we present LoFi, a medical vision foundation model built on location-aware fine-grained representation learning. LoFi trains a vision encoder with a lightweight large language model under grounding and grounded captioning objectives. Because these objectives require predicting location from clinical text and vice versa, spatial consistency emerges without any explicit patch-level regularization. To enable training at scale, we construct MedG, a large-scale medical grounding dataset of 4.48M image-text-box triplets curated from 84 datasets spanning 7 modalities. Across phrase grounding, visual question answering, and region-based organ classification under perturbations, LoFi consistently outperforms general-purpose and medical vision foundation models as well as state-of-the-art LVLMs. Code is available at this https URL.
Abstract (translated)
URL
https://arxiv.org/abs/2608.00976
PDF
-
SAGE: An Expert-Annotated South Asian GI Endoscopy Dataset for Multimodal Learning and Hallucination Analysis
2026-08-01 18:24:37Niyoj Oli, Sachin Acharya, Sandesh Pokhrel, Sanjay Bhandari, Ramesh Rana, Nikesh Mani Shrestha, Ram Bahadur Gurung, Yash Raj Shrestha, Prashnna K Gyawali, Binod BhattaraiAbstract
Gastrointestinal cancers represent a growing health burden in the South Asian region, driven largely by rapid changes in socio-economic conditions and lifestyle habits. However, early diagnosis remains limited by inadequate equipment, financial resources, and scarce GI expertise. AI-assisted diagnosis and report generation show great promise in alleviating this problem by providing non-specialist healthcare workers the technical expertise to perform diagnosis. Yet, almost all open-source, publicly available datasets are predominantly collected from the European region, with minimal representation from the South Asian region. The lack of open-source GI datasets from diverse geographic regions has made it difficult to assess whether population bias is present in existing models, and to develop geographically inclusive AI tools for automated GI diagnosis. To address this gap, we introduce SAGE: An Expert-Annotated South Asian GI Endoscopy dataset for Multimodal Learning and Hallucination Analysis, for image captioning, multi-label classification, and VQA tasks. It consists of 1,300 images, their captions along with hallucination tag, 18 labels and 14,276 question-answer pairs making it well-suited for diverse range of tasks including classification, benchmarking, and fine-tuning LMMs. We further conducted benchmarking of task-specific models, like multi-class classifiers on the effect of population shift which reveals that such models suffers the most with an average F1 score drop of 54 points on South Asian dataset. Further, benchmarking of contemporary LMMs reveals a substantial drop in the average GREEN score for anatomical landmark detection (0.308) and abnormality detection (0.410). We open-source our dataset under the CC BY-SA 4.0 license, and we hope to encourage others to contribute toward more inclusive dataset representation and help counteract population bias in medical AI.
Abstract (translated)
URL
https://arxiv.org/abs/2606.22144
PDF
-
LUT: Latent Utility Training for Visual Reasoning
2026-08-01 16:27:45Jiaxuan Kang, Siyu Chen, Mingda Li, Mingjie Liu, Tianyue Wang, Zhaoyang Wei, Yongheng Zhang, Yanchao Hao, Zheng WeiAbstract
Multimodal large language models have advanced visual understanding, yet perception-intensive reasoning remains challenging. Recent latent visual reasoning methods introduce hidden-space computation before answering, but they often rely on costly intermediate supervision, such as bounding boxes, sketches, or interleaved rationales. These strategies focus on how latent states should be shaped, but do not explicitly assess whether the latent is useful for the final answer. We propose LUT, a latent reasoning framework trained with only standard VQA pairs. LUT centers training on Latent Utility at two levels. At the trajectory level, we propose Utility-Aware Latent Distillation SFT, which explores answer-relevant latent trajectories, selects qualified trajectories by their information gain, and distills more reliable and learnable supervision through curriculum learning. At the step level, we propose Latent Attribution Policy Optimization, which uses answer-to-latent attribution to differentially optimize latent steps during reinforcement learning. Experiments on perception-intensive visual reasoning benchmarks show that LUT outperforms previous latent reasoning methods and remains competitive with latent-text interleaved methods with lower annotation cost.
Abstract (translated)
URL
https://arxiv.org/abs/2608.00743
PDF
