Think Twice to See More: Iterative Visual Reasoning in Medical VLMs

• URL: http://arxiv.org/abs/2510.10052v1
• Date: Sat, 11 Oct 2025 06:39:57 GMT
• Title: Think Twice to See More: Iterative Visual Reasoning in Medical VLMs
• Authors: Kaitao Chen, Shaohao Rui, Yankai Jiang, Jiamin Wu, Qihao Zheng, Chunfeng Song, Xiaosong Wang, Mu Zhou, Mianxin Liu,
• Abstract summary: We introduce ViTAR, a framework that emulates the iterative reasoning process of human experts through a cognitive chain of "think-act-rethink-answer"<n>ViTAR treats medical images as interactive objects, enabling models to engage multi-step visual reasoning.
• Score: 21.083636394814217
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Medical vision-language models (VLMs) excel at image-text understanding but typically rely on a single-pass reasoning that neglects localized visual cues. In clinical practice, however, human experts iteratively scan, focus, and refine the regions of interest before reaching a final diagnosis. To narrow this machine-human perception gap, we introduce ViTAR, a novel VLM framework that emulates the iterative reasoning process of human experts through a cognitive chain of "think-act-rethink-answer". ViTAR treats medical images as interactive objects, enabling models to engage multi-step visual reasoning. To support this approach, we curate a high-quality instruction dataset comprising 1K interactive examples that encode expert-like diagnostic behaviors. In addition, a 16K visual question answering training data has been curated towards fine-grained visual diagnosis. We introduce a two-stage training strategy that begins with supervised fine-tuning to guide cognitive trajectories, followed by the reinforcement learning to optimize decision-making. Extensive evaluations demonstrate that ViTAR outperforms strong state-of-the-art models. Visual attention analysis reveals that from the "think" to "rethink" rounds, ViTAR increasingly anchors visual grounding to clinically critical regions and maintains high attention allocation to visual tokens during reasoning, providing mechanistic insight into its improved performance. These findings demonstrate that embedding expert-style iterative thinking chains into VLMs enhances both performance and trustworthiness of medical AI.

Related papers

• ViThinker: Active Vision-Language Reasoning via Dynamic Perceptual Querying [15.728211622542267]
  ViThinker is a framework that enables vision-language models to autonomously generate decision tokens triggering the synthesis of expert-aligned visual features on demand.<n>ViThinker internalizes vision-expert capabilities during training, performing generative mental simulation during inference without external tool calls.
  arXiv  Detail & Related papers  (2026-02-02T22:29:57Z)
• Toward Cognitive Supersensing in Multimodal Large Language Model [67.15559571626747]
  We introduce Cognitive Supersensing, a training paradigm that endows MLLMs with human-like visual imagery capabilities.<n>In experiments, MLLMs trained with Cognitive Supersensing significantly outperform state-of-the-art baselines on CogSense-Bench.<n>We will open-source the CogSense-Bench and our model weights.
  arXiv  Detail & Related papers  (2026-02-02T02:19:50Z)
• MedEyes: Learning Dynamic Visual Focus for Medical Progressive Diagnosis [17.59077756990045]
  MedEyes is a reinforcement learning framework that dynamically models clinician-style diagnostic reasoning.<n>It emulates the diagnostic process through a dual-mode exploration strategy, scanning for systematic abnormality localization and drilling for detailed regional analysis.<n>Experiments demonstrate that MedEyes achieves an average performance improvement of +8.5% across multiple medical VQA benchmarks.
  arXiv  Detail & Related papers  (2025-11-27T01:47:43Z)
• S-Chain: Structured Visual Chain-of-Thought For Medicine [81.97605645734741]
  We introduce S-Chain, the first large-scale dataset of 12,000 expert-annotated medical images with bounding boxes and structured visual CoT (SV-CoT)<n>The dataset further supports 16 languages, totaling over 700k VQA pairs for broad multilingual applicability.<n>S-Chain establishes a new benchmark for grounded medical reasoning and paves the way toward more trustworthy and explainable medical vision-language models.
  arXiv  Detail & Related papers  (2025-10-26T15:57:14Z)
• MedAlign: A Synergistic Framework of Multimodal Preference Optimization and Federated Meta-Cognitive Reasoning [52.064286116035134]
  We develop MedAlign, a framework to ensure visually accurate LVLM responses for Medical Visual Question Answering (Med-VQA)<n>We first propose a multimodal Direct Preference Optimization (mDPO) objective to align preference learning with visual context.<n>We then design a Retrieval-Aware Mixture-of-Experts (RA-MoE) architecture that utilizes image and text similarity to route queries to a specialized and context-augmented LVLM.
  arXiv  Detail & Related papers  (2025-10-24T02:11:05Z)
• Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning [79.34909830834464]
  Vision-Language Models (VLMs) have demonstrated remarkable success across diverse visual tasks, yet their performance degrades in complex visual environments.<n>We show that visual complexity strongly correlates with attention entropy, negatively impacting reasoning performance.<n>We propose Contrastive Attention Refinement for Visual Enhancement (CARVE), a training-free method that extracts task-relevant visual signals through attention contrasting at the pixel level.
  arXiv  Detail & Related papers  (2025-09-08T09:20:04Z)
• GEMeX-ThinkVG: Towards Thinking with Visual Grounding in Medical VQA via Reinforcement Learning [50.94508930739623]
  Medical visual question answering aims to support clinical decision-making by enabling models to answer natural language questions based on medical images.<n>Current methods still suffer from limited answer reliability and poor interpretability, impairing the ability of clinicians and patients to understand and trust model-generated answers.<n>This work first proposes a Thinking with Visual Grounding dataset wherein the answer generation is decomposed into intermediate reasoning steps.<n>We introduce a novel verifiable reward mechanism for reinforcement learning to guide post-training, improving the alignment between the model's reasoning process and its final answer.
  arXiv  Detail & Related papers  (2025-06-22T08:09:58Z)
• VisionReasoner: Unified Reasoning-Integrated Visual Perception via Reinforcement Learning [56.99825489208698]
  We introduce VisionReasoner, a unified framework capable of reasoning and solving multiple visual perception tasks.<n> VisionReasoner enhances its reasoning capabilities to analyze visual inputs, and addresses diverse perception tasks within a unified model.<n>We evaluate VisionReasoner on ten diverse tasks spanning three critical domains: detection, segmentation, and counting.
  arXiv  Detail & Related papers  (2025-05-17T16:51:47Z)
• A Cognitive Paradigm Approach to Probe the Perception-Reasoning Interface in VLMs [3.2228025627337864]
  This paper introduces a structured evaluation framework to dissect the perception-reasoning interface in Vision-Language Models (VLMs)<n>We propose three distinct evaluation paradigms, mirroring human problem-solving strategies.<n>Applying this framework, we demonstrate that CA, leveraging powerful language models for reasoning over rich, independently generated descriptions, achieves new state-of-the-art (SOTA) performance.
  arXiv  Detail & Related papers  (2025-01-23T12:42:42Z)
• Guiding Medical Vision-Language Models with Explicit Visual Prompts: Framework Design and Comprehensive Exploration of Prompt Variations [15.052986179046076]
  We introduce MedVP, a pioneering framework that integrates medical entity extraction, visual prompt generation, and dataset adaptation for visual prompt guided fine-tuning.<n>We successfully outperform recent state-of-the-art large models across multiple medical VQA datasets.
  arXiv  Detail & Related papers  (2025-01-04T21:23:36Z)
• Cracking the Code of Hallucination in LVLMs with Vision-aware Head Divergence [69.86946427928511]
  We investigate the internal mechanisms driving hallucination in large vision-language models (LVLMs)<n>We introduce Vision-aware Head Divergence (VHD), a metric that quantifies the sensitivity of attention head outputs to visual context.<n>We propose Vision-aware Head Reinforcement (VHR), a training-free approach to mitigate hallucination by enhancing the role of vision-aware attention heads.
  arXiv  Detail & Related papers  (2024-12-18T15:29:30Z)
• UnICLAM:Contrastive Representation Learning with Adversarial Masking for Unified and Interpretable Medical Vision Question Answering [7.2486693553383805]
  Current Medical-VQA models learn cross-modal representations through residing vision and texture encoders in dual separate spaces. We propose UnICLAM, a Unified and Interpretable Medical-VQA model through Contrastive Representation Learning with Adversarial Masking. Experimental results on VQA-RAD and SLAKE public benchmarks demonstrate that UnICLAM outperforms existing 11 state-of-the-art Medical-VQA models.
  arXiv  Detail & Related papers  (2022-12-21T02:48:15Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.