Pith. sign in

REVIEW 19 cited by

MedVLM-R1: Incentivizing Medical Reasoning Capability of Vision-Language Models (VLMs) via Reinforcement Learning

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2502.19634 v2 pith:KAM75S3R submitted 2025-02-26 cs.CV cs.AI

Jiazhen Pan , Che Liu , Junde Wu , Fenglin Liu , Jiayuan Zhu , Hongwei Bran Li , Chen Chen , Cheng Ouyang
show 1 more author
Daniel Rueckert
This is my paper
classification cs.CVcs.AI
keywords reasoningmedvlm-r1medicalmodelmodelsvlmsanalysisimage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

Reasoning is a critical frontier for advancing medical image analysis, where transparency and trustworthiness play a central role in both clinician trust and regulatory approval. Although Medical Visual Language Models (VLMs) show promise for radiological tasks, most existing VLMs merely produce final answers without revealing the underlying reasoning. To address this gap, we introduce MedVLM-R1, a medical VLM that explicitly generates natural language reasoning to enhance transparency and trustworthiness. Instead of relying on supervised fine-tuning (SFT), which often suffers from overfitting to training distributions and fails to foster genuine reasoning, MedVLM-R1 employs a reinforcement learning framework that incentivizes the model to discover human-interpretable reasoning paths without using any reasoning references. Despite limited training data (600 visual question answering samples) and model parameters (2B), MedVLM-R1 boosts accuracy from 55.11% to 78.22% across MRI, CT, and X-ray benchmarks, outperforming larger models trained on over a million samples. It also demonstrates robust domain generalization under out-of-distribution tasks. By unifying medical image analysis with explicit reasoning, MedVLM-R1 marks a pivotal step toward trustworthy and interpretable AI in clinical practice. Inference model is available at: https://huggingface.co/JZPeterPan/MedVLM-R1.

Discussion (0). Sign in to comment.

Forward citations

Cited by 19 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. Beyond Medical Diagnostics: How Medical Multimodal Large Language Models Think in Space

    cs.CV 2026-03 conditional novelty 6.0 of 10

    SpatialMed provides the first CT-based benchmark of 3D spatial reasoning for medical MLLMs, on which 14 models perform near chance, particularly for distance and volume estimation.

  2. Towards Better Dental AI: A Multimodal Benchmark and Instruction Dataset for Panoramic X-ray Analysis

    cs.CV 2025-09 reject novelty 6.0 of 10

    MMOral is a large new dental X-ray instruction dataset and benchmark, but the proposed model's 24.73% improvement is from fine-tuning and then testing on the same data pool.

  3. Addressing Benchmarking Gaps in Large Language Models for Health and Medicine with Dynamic Red-Teaming

    cs.LG 2025-07 reject novelty 6.0 of 10

    A dynamic red-teaming audit reports that 94% of MedQA-correct answers fail under adversarial mutation, with 86% privacy leak rates, 81% bias shift rates, and 66-74% hallucination rates across 15 medical LLMs.

  4. Unleashing Embodied Task Planning Ability in LLMs via Reinforcement Learning

    cs.CL 2025-06 conditional novelty 6.0 of 10

    A 7B LLM trained with sparse completion rewards and a GRPO-style algorithm reaches state-of-the-art on ALFWorld and ScienceWorld.

  5. ComfyUI-R1: Exploring Reasoning Models for Workflow Generation

    cs.CL 2025-06 conditional novelty 6.0 of 10

    A 7B reasoning model trained with supervised fine-tuning and reinforcement learning generates ComfyUI workflows from text instructions and outperforms GPT-4o and Claude-based baselines on the authors' tests.

  6. Interpretable and Reliable Detection of AI-Generated Images via Grounded Reasoning in MLLMs

    cs.CV 2025-06 conditional novelty 6.0 of 10

    Fine-tuning Qwen-2.5-VL on the new FakeXplained dataset of 8,772 AI-generated images with box-and-caption artifact annotations yields an explainable detector with 98.1% accuracy and 37.8% IoU.

  7. Q-Ponder: A Unified Training Pipeline for Reasoning-based Visual Quality Assessment

    cs.CV 2025-06 conditional novelty 6.0 of 10

    Q-Ponder is a two-stage pipeline (distill-then-reinforce) that makes a 7B multimodal model both more accurate at image quality scoring and better at explaining its judgments.

  8. MedBookVQA: A Systematic and Comprehensive Medical Benchmark Derived from Open-Access Book

    cs.AI 2025-06 conditional novelty 6.0 of 10

    MedBookVQA is a new 5,000-question, textbook-derived multimodal benchmark for testing medical AI systems, with labels for imaging modality, body anatomy, and clinical specialty.

  9. Training LLMs for EHR-Based Reasoning Tasks via Reinforcement Learning

    cs.CL 2025-05 conditional novelty 6.0 of 10

    A 3B model trained with a small SFT warm-up followed by verifiable-reward RL matches or exceeds far larger models on EHR-based medical calculation, trial matching, and diagnosis tasks.

  10. Collision- and Reachability-Aware Multi-Robot Control with Grounded LLM Planners

    cs.RO 2025-05 conditional novelty 6.0 of 10

    RLVR fine-tuning teaches small LLMs to reason about reachability and collisions, letting them beat far larger ungrounded LLMs on multi-robot box-moving tasks.

  11. CX-Mind: A Pioneering Multimodal Large Language Model for Interleaved Reasoning in Chest X-ray via Curriculum-Guided Reinforcement Learning

    cs.LG 2025-07 conditional novelty 5.0 of 10

    CX-Mind combines curriculum reinforcement learning and rule-based process rewards to train a chest X-ray vision-language model that produces interleaved think-answer reasoning and reports state-of-the-art results acro...

  12. RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints

    cs.CV 2025-06 reject novelty 5.0 of 10

    RARL fine-tunes Qwen2-VL-2B on 716 medical samples with GRPO, LoRA, and a vaguely defined reasoning reward, claiming gains of 7.78% over SFT on reasoning and up to 27% on unseen VQA benchmarks.

  13. Objective-Aligned Direct Answer SFT for Robust Multi-Frame Medical VQA

    cs.CV 2026-07 conditional novelty 4.0 of 10

    Direct answer-only supervised fine-tuning is the most robust adaptation family on MedFrameQA, beating frozen baselines by ~6 points and outperforming complex variants on seed stability.

  14. MedFoundationHub: A Lightweight and Secure Toolkit for Deploying Medical Vision Language Foundation Models

    cs.CV 2025-08 conditional novelty 4.0 of 10

    A GUI toolkit for secure local deployment of medical vision-language models, with a pathologist-scored evaluation showing these models still frequently fail on pathology questions.

  15. The Philosophy and Physics of Duality

    physics.hist-ph 2025-08 unverdicted novelty 4.0 of 10

    A philosophical monograph that surveys dualities across physics and proposes a 'geometric view of theories' for theoretical equivalence, realism, and explanation.

  16. How Far Have Medical Vision-Language Models Come? A Comprehensive Benchmarking Study

    cs.CV 2025-07 reject novelty 4.0 of 10

    A ten-model, seven-benchmark medical VLM evaluation whose headline reasoning-vs-understanding finding is contradicted by its own tables.

  17. Improving Medical Reasoning with Curriculum-Aware Reinforcement Learning

    cs.AI 2025-05 conditional novelty 4.0 of 10

    A curriculum-based GRPO schedule that first trains on close-ended medical VQA and then on open-ended VQA improves benchmark scores over joint training and vanilla RL, though the open-ended metric is the training objec...

  18. Beyond Distillation: Pushing the Limits of Medical LLM Reasoning with Minimalist Rule-Based RL

    cs.CL 2025-05 conditional novelty 4.0 of 10

    Medical QA accuracy improves substantially through reinforcement learning with a binary correct-answer reward alone, without supervised fine-tuning on distilled reasoning traces.

  19. Reinforcement Fine-Tuning Powers Reasoning Capability of Multimodal Large Language Models

    cs.CL 2025-05 conditional novelty 2.0 of 10

    A survey-style position paper claims that reinforcement fine-tuning powers reasoning in multimodal LLMs, summarizing over a hundred recent works and proposing five future research directions.

Pith tools