When Does RL Help Medical VLMs? Disentangling Vision, SFT, and RL Gains

• URL: http://arxiv.org/abs/2603.01301v1
• Date: Sun, 01 Mar 2026 22:16:19 GMT
• Title: When Does RL Help Medical VLMs? Disentangling Vision, SFT, and RL Gains
• Authors: Ahmadreza Jeddi, Kimia Shaban, Negin Baghbanzadeh, Natasha Sharan, Abhishek Moturu, Elham Dolatabadi, Babak Taati,
• Abstract summary: Reinforcement learning (RL) is increasingly used to post-train medical Vision-Language Models (VLMs)<n>It remains unclear whether RL improves medical visual reasoning or mainly sharpens behaviors already induced by supervised fine-tuning (SFT)<n>We present a controlled study that disentangles these effects along three axes: vision, SFT, and RL.
• Score: 1.9256950761509062
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning (RL) is increasingly used to post-train medical Vision-Language Models (VLMs), yet it remains unclear whether RL improves medical visual reasoning or mainly sharpens behaviors already induced by supervised fine-tuning (SFT). We present a controlled study that disentangles these effects along three axes: vision, SFT, and RL. Using MedMNIST as a multi-modality testbed, we probe visual perception by benchmarking VLM vision towers against vision-only baselines, quantify reasoning support and sampling efficiency via Accuracy@1 versus Pass@K, and evaluate when RL closes the support gap and how gains transfer across modalities. We find that RL is most effective when the model already has non-trivial support (high Pass@K): it primarily sharpens the output distribution, improving Acc@1 and sampling efficiency, while SFT expands support and makes RL effective. Based on these findings, we propose a boundary-aware recipe and instantiate it by RL post-training an OctoMed-initialized model on a small, balanced subset of PMC multiple-choice VQA, achieving strong average performance across six medical VQA benchmarks.

Related papers

• CVeDRL: An Efficient Code Verifier via Difficulty-aware Reinforcement Learning [57.24524263804788]
  Code verifiers play a critical role in post-verification for LLM-based code generation.<n>Existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency.<n>We show that naive RL with only functionality rewards fails to generate effective unit tests for difficult branches and samples.
  arXiv  Detail & Related papers  (2026-01-30T10:33:29Z)
• Benchmark Success, Clinical Failure: When Reinforcement Learning Optimizes for Benchmarks, Not Patients [2.377303603725137]
  We introduce ChexReason, a vision-language model trained via R1-style methodology (SFT followed by GRPO) using only 2,000 SFT samples, 1,000 RL samples, and a single A100 GPU.<n> Evaluations on CheXpert and NIH benchmarks reveal a fundamental tension: GRPO recovers in-distribution performance (23% improvement on CheXpert, macro-F1 = 0.346) but degrades cross-dataset transferability (19% drop on NIH)<n>We identify a generalization paradox where the SFT checkpoint uniquely improves on NIH before optimization, indicating teacher-guided reasoning captures more institution-agnostic features.
  arXiv  Detail & Related papers  (2025-12-28T21:57:42Z)
• Reassessing the Role of Supervised Fine-Tuning: An Empirical Study in VLM Reasoning [30.751908700207185]
  SFT plays a crucial role across several scenarios.<n>SFT with only 2K achieves comparable or better reasoning performance to RL with 20K.<n>We identify a pervasive issue of deceptive rewards, where higher rewards fail to correlate with better reasoning accuracy in RL.
  arXiv  Detail & Related papers  (2025-12-14T13:46:42Z)
• More Than the Final Answer: Improving Visual Extraction and Logical Consistency in Vision-Language Models [74.10138874771852]
  We propose PeRL-VL (Perception and Reasoning Learning for Vision-Language Models), a decoupled framework that separately improves visual perception and textual reasoning on top of RLVR.<n>For perception, PeRL-VL introduces a VLM-based description reward that scores the model's self-generated image descriptions for faithfulness and sufficiency.<n>For reasoning, PeRL-VL adds a text-only Reasoning SFT stage on logic-rich chain-of-thought data, enhancing coherence and logical consistency independently of vision.
  arXiv  Detail & Related papers  (2025-12-13T23:06:18Z)
• Enhancing Radiology Report Generation and Visual Grounding using Reinforcement Learning [15.894854593567963]
  Reinforcement learning can incorporate task-specific feedback, and its combination with explicit intermediate reasoning ("thinking") has demonstrated substantial gains on verifiable math and coding tasks.<n>We build an updated vision-language model based on Qwen3-VL, followed by a cold-start SFT stage that equips the model with basic thinking ability.<n>We find that while strong SFT remains crucial for high base performance, RL provides additional gains on both tasks, whereas explicit thinking does not appear to further improve results.
  arXiv  Detail & Related papers  (2025-12-11T14:36:14Z)
• RL makes MLLMs see better than SFT [96.508432109136]
  We conduct a critical yet under-explored analysis of the vision encoder of Multimodal Language Model (MLLM)<n>Our results demonstrate that MLLM's post-training strategy (i.e., SFT or RL) not only leads to distinct outcomes on MLLM downstream tasks, but also fundamentally reshapes MLLM's underlying visual representations.<n>We then reframe our findings into a simple recipe for building strong vision encoders for MLLMs, Preference-Instructed Vision OpTimization (PIVOT)
  arXiv  Detail & Related papers  (2025-10-18T03:37:17Z)
• RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints [0.0]
  Reasoning-Aware Reinforcement Learning framework enhances the reasoning capabilities of medical vision-language models.<n>Our approach fine-tunes a lightweight base model, Qwen2-VL-2B-Instruct, using Low-Rank Adaptation and custom reward functions.<n> Experimental results show RARL significantly improves VLM performance in medical image analysis and clinical reasoning.
  arXiv  Detail & Related papers  (2025-06-07T00:26:23Z)
• Beyond Accuracy: Dissecting Mathematical Reasoning for LLMs Under Reinforcement Learning [93.00629872970364]
  Reinforcement learning (RL) has become the dominant paradigm for improving the performance of language models on complex reasoning tasks.<n>We introduce SPARKLE, a fine-grained analytic framework to dissect the effects of RL across three key dimensions.<n>We study whether difficult problems -- those yielding no RL signals and mixed-quality reasoning traces -- can still be effectively used for training.
  arXiv  Detail & Related papers  (2025-06-05T07:53:59Z)
• SFT or RL? An Early Investigation into Training R1-Like Reasoning Large Vision-Language Models [39.551767637896404]
  This work revisits the dominant supervised fine-tuning (SFT) then reinforcement learning (RL) paradigm for training Large Vision-Language Models (LVLMs)<n>We show that SFT can significantly undermine subsequent RL by inducing pseudo reasoning paths'' imitated from expert models.<n>We introduce VLAA-Thinking, a new multimodal dataset designed to support reasoning in LVLMs.
  arXiv  Detail & Related papers  (2025-04-10T16:54:05Z)
• Exploring the Effect of Reinforcement Learning on Video Understanding: Insights from SEED-Bench-R1 [53.894789613838654]
  We introduce SEED-Bench-R1, a benchmark designed to evaluate post-training methods for MLLMs in video understanding.<n>It includes intricate real-world videos and complex everyday planning tasks in the format of multiple-choice questions.<n>Using Qwen2-VL-Instruct-7B as a base model, we compare RL with supervised fine-tuning (SFT)<n>Our detailed analysis reveals that RL enhances visual perception but often produces less coherent reasoning chains.
  arXiv  Detail & Related papers  (2025-03-31T17:55:23Z)
• OpenVLThinker: Complex Vision-Language Reasoning via Iterative SFT-RL Cycles [91.88062410741833]
  We introduce OpenVLThinker, one of the first open-source large vision-language models (LVLMs) to exhibit sophisticated chain-of-thought reasoning.<n>We show that OpenVLThinker-7B consistently advances performance across six benchmarks demanding mathematical and general reasoning.
  arXiv  Detail & Related papers  (2025-03-21T17:52:43Z)
• Think or Not Think: A Study of Explicit Thinking in Rule-Based Visual Reinforcement Fine-Tuning [26.835266813794316]
  We first propose CLS-RL for MLLM image classification, using verifiable rewards for fine-tuning.<n>We then rethink and question whether explicit thinking in RFT is always necessary.<n>No-Thinking-RL explores RFT without thinking by introducing a simple equality accuracy reward.
  arXiv  Detail & Related papers  (2025-03-20T14:37:45Z)

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.