MEDVISTAGYM: A Scalable Training Environment for Thinking with Medical Images via Tool-Integrated Reinforcement Learning
This addresses the challenge of enabling effective tool-integrated reasoning for medical image analysis, which is incremental as it builds on existing tool-augmented methods with a new training approach.
The paper tackles the problem of vision language models struggling with multi-step reasoning on medical images by introducing MedVistaGym, a scalable training environment for tool-integrated reinforcement learning, resulting in MedVistaGym-R1-8B outperforming baselines by 19.10% to 24.21% on six medical VQA benchmarks.
Vision language models (VLMs) achieve strong performance on general image understanding but struggle to think with medical images, especially when performing multi-step reasoning through iterative visual interaction. Medical VLMs often rely on static visual embeddings and single-pass inference, preventing models from re-examining, verifying, or refining visual evidence during reasoning. While tool-integrated reasoning offers a promising path forward, open-source VLMs lack the training infrastructure to learn effective tool selection, invocation, and coordination in multi-modal medical reasoning. We introduce MedVistaGym, a scalable and interactive training environment that incentivizes tool-integrated visual reasoning for medical image analysis. MedVistaGym equips VLMs to determine when and which tools to invoke, localize task-relevant image regions, and integrate single or multiple sub-image evidence into interleaved multimodal reasoning within a unified, executable interface for agentic training. Using MedVistaGym, we train MedVistaGym-R1 to interleave tool use with agentic reasoning through trajectory sampling and end-to-end reinforcement learning. Across six medical VQA benchmarks, MedVistaGym-R1-8B exceeds comparably sized tool-augmented baselines by 19.10% to 24.21%, demonstrating that structured agentic training--not tool access alone--unlocks effective tool-integrated reasoning for medical image analysis.