# MedEyes: Learning Dynamic Visual Focus for Medical Progressive Diagnosis

arXiv: https://arxiv.org/abs/2511.22018 · HTML: https://mlsanity.com/paper/2511.22018/

## Key facts

- Cited by 32 papers in the map (data through 14 Oct 2026).
- Placed in Clinical and biomedical applications, one of 17 directions in LLM reasoning.
- Builds on 12 of its 23 references that are in the map.
- First posted on arXiv 27 Nov 2025.

## Placed in

- LLM reasoning / [Clinical and biomedical applications](https://mlsanity.com/lr/clinical-and-biomedical-applications/)

## Builds on (most cited first, 12 of 12)

- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [The Curious Case of Neural Text Degeneration](https://mlsanity.com/paper/1904.09751/) (2019)
- [InternVL: Scaling up Vision Foundation Models and Aligning for Generic Visual-Linguistic Tasks](https://mlsanity.com/paper/2312.14238/) (2023)
- [MMMU: A Massive Multi-discipline Multimodal Understanding and Reasoning Benchmark for Expert AGI](https://mlsanity.com/paper/2311.16502/) (2023)
- [LLaVA-Med: Training a Large Language-and-Vision Assistant for Biomedicine in One Day](https://mlsanity.com/paper/2306.00890/) (2023)
- [PathVQA: 30000+ Questions for Medical Visual Question Answering](https://mlsanity.com/paper/2003.10286/) (2020)
- [SLAKE: A Semantically-Labeled Knowledge-Enhanced Dataset for Medical Visual Question Answering](https://mlsanity.com/paper/2102.09542/) (2021)
- [Med-Flamingo: a Multimodal Medical Few-shot Learner](https://mlsanity.com/paper/2307.15189/) (2023)
- [GRIT: Teaching MLLMs to Think with Images](https://mlsanity.com/paper/2505.15879/) (2025)
- [Towards a Multimodal Large Language Model with Pixel-Level Insight for Biomedicine](https://mlsanity.com/paper/2412.09278/) (2024)
- [GMAI-VL & GMAI-VL-5.5M: A Large Vision-Language Model and A Comprehensive Multimodal Dataset Towards General Medical AI](https://mlsanity.com/paper/2411.14522/) (2024)
- [Deep semantic gaze embedding and scanpath comparison for expertise classification during OPT viewing](https://mlsanity.com/paper/2003.13987/) (2020)

## Built on by (most cited first, 30 of 32)

- [GaussianDWM: 3D Gaussian Driving World Model for Unified Scene Understanding and Multi-Modal Generation](https://mlsanity.com/paper/2512.23180/) (2025)
- [Med-CMR: A Fine-Grained Benchmark Integrating Visual Evidence and Clinical Logic for Medical Complex Multimodal Reasoning](https://mlsanity.com/paper/2512.00818/) (2025)
- [Real-Time Aligned Reward Model beyond Semantics](https://mlsanity.com/paper/2601.22664/) (2026)
- [MedSynapse-V: Bridging Visual Perception and Clinical Intuition via Latent Memory Evolution](https://mlsanity.com/paper/2604.26283/) (2026)
- [ChainCaps: Composition-Safe Tool-Using Agents via Monotonic Capability Attenuation](https://mlsanity.com/paper/2605.26542/) (2026)
- [Recursive Self-Evolving Agents via Held-Out Selection](https://mlsanity.com/paper/2606.28374/) (2026)
- [SpatialReward: Verifiable Spatial Reward Modeling for Fine-Grained Spatial Consistency in Text-to-Image Generation](https://mlsanity.com/paper/2603.22228/) (2026)
- [One Refiner to Unlock Them All: Inference-Time Reasoning Elicitation via Reinforcement Query Refinement](https://mlsanity.com/paper/2604.25444/) (2026)
- [CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal Reasoning](https://mlsanity.com/paper/2512.19554/) (2025)
- [Tracking Drift: Variation-Aware Entropy Scheduling for Non-Stationary Reinforcement Learning](https://mlsanity.com/paper/2601.19624/) (2026)
- [DyJR: Preserving Diversity in Reinforcement Learning with Verifiable Rewards via Dynamic Jensen-Shannon Replay](https://mlsanity.com/paper/2603.16157/) (2026)
- [From Internal Diagnosis to External Auditing: A VLM-Driven Paradigm for Data-Free Online Backdoor Defense](https://mlsanity.com/paper/2601.19448/) (2026)
- [SQL-ASTRA: Alleviating Sparse Feedback in Agentic SQL via Column-Set Matching and Trajectory Aggregation](https://mlsanity.com/paper/2603.16161/) (2026)
- [VistaHop: Benchmarking Long-Horizon Visual DeepSearch](https://mlsanity.com/paper/2606.03273/) (2026)
- [Learning Transferable Temporal Primitives for Video Reasoning via Synthetic Videos](https://mlsanity.com/paper/2603.17693/) (2026)
- [Thinking in Scales: Accelerating Gigapixel Pathology Image Analysis via Adaptive Continuous Reasoning](https://mlsanity.com/paper/2605.19491/) (2026)
- [Deformba: Vision State Space Model with Adaptive State Fusion](https://mlsanity.com/paper/2605.21308/) (2026)
- [MM-Snowball: Evaluating and Mitigating Hallucination Snowballing in Multimodal Multi-Turn Dialogue](https://mlsanity.com/paper/2606.00622/) (2026)
- [TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation](https://mlsanity.com/paper/2606.07688/) (2026)
- [REFLEX: Reflective Evolution from LLM Experience](https://mlsanity.com/paper/2606.16496/) (2026)
- [Dual Distribution Estimation for Zero-shot Noisy Test-Time Adaptation with VLMs](https://mlsanity.com/paper/2606.25758/) (2026)
- [The Path to Self-Evolving Clinical Systems: Scaling Medical Agents from Assistance to Autonomy](https://mlsanity.com/paper/2607.11175/) (2026)
- [All for 1-Bit: Towards Genuine 1-Bit Post-Training Quantization for LLMs](https://mlsanity.com/paper/2609.06161/) (2026)
- [Exploring Accurate and Transparent Domain Adaptation in Predictive Healthcare via Concept-Grounded Orthogonal Inference](https://mlsanity.com/paper/2602.12542/) (2026)
- [Dynamic Model Merging Made Slim](https://mlsanity.com/paper/2605.18904/) (2026)
- [Modality-Decoupled Online Recursive Editing](https://mlsanity.com/paper/2605.20273/) (2026)
- [Skill Weaving: Efficient LLM Improvement via Modular Skillpacks](https://mlsanity.com/paper/2605.22205/) (2026)
- [CRAFT: Causal Responsibility and Failure Tracing in Medical Vision Language Models](https://mlsanity.com/paper/2609.38810/) (2026)
- [Shift-Dependent Asymmetry: Orthogonal Inverse Low-Rank Adaptation for Federated Medical Segmentation](https://mlsanity.com/paper/2606.08687/) (2026)
- [Two-Stage Cross-Domain Cervical Abnormality Screening with Cytopathological Image Synthesis and Knowledge Distillation](https://mlsanity.com/paper/2606.27678/) (2026)

## Abstract (arXiv)

Accurate medical diagnosis often involves progressive visual focusing and iterative reasoning, characteristics commonly observed in clinical workflows. While recent vision-language models demonstrate promising chain-of-thought (CoT) reasoning capabilities via reinforcement learning with verifiable rewards (RLVR), their purely on-policy learning paradigm tends to reinforce superficially coherent but clinically inaccurate reasoning paths. We propose MedEyes, a novel reinforcement learning framework that dynamically models clinician-style diagnostic reasoning by progressively attending to and interpreting relevant medical image regions. By incorporating off-policy expert guidance, MedEyes converts expert visual search trajectories into structured external behavioral signals, guiding the model toward clinically aligned visual reasoning. We design the Gaze-guided Reasoning Navigator (GRN) to emulate the diagnostic process through a dual-mode exploration strategy, scanning for systematic abnormality localization and drilling for detailed regional analysis. To balance expert imitation and autonomous discovery, we introduce the Confidence Value Sampler (CVS), which employs nucleus sampling and adaptive termination to create diverse yet credible exploration paths. Finally, the dual-stream GRPO optimization framework decouples on-policy and off-policy learning signals, mitigating reward assimilation and entropy collapse. Experiments demonstrate that MedEyes achieves an average performance improvement of +8.5pp across multiple medical VQA benchmarks, validating MedEyes's potential in building trustworthy medical AI systems. Code is available at https://github.com/zhcz328/MedEyes.