arXiv 2601.071072026-01-12cs.CVcs.AI

MEDVISTAGYM：一种通过工具集成强化学习进行医学图像思考的可扩展训练环境

MEDVISTAGYM: A Scalable Training Environment for Thinking with Medical Images via Tool-Integrated Reinforcement Learning

Virginia Tech（弗吉尼亚理工大学）；UT Southwestern Medical Center（德克萨斯大学西南医学中心）；Georgia Institute of Technology（佐治亚理工学院）；Cisco（思科公司）

Meng Lu, Yuxing Lu, Yuchen Zhuang, Megan Mullins, Yang Xie, Guanghua Xiao, Charles Fleming, Wenqi Shi, Xuan Wang

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

MedVistaGym通过工具集成强化学习提升医学图像分析的代理训练效果。

摘要：

视觉语言模型（VLMs）在一般图像理解上表现强劲，但在处理医学图像时尤其在需要通过迭代视觉交互进行多步推理时表现出困难。医学VLMs通常依赖静态视觉嵌入和单次推理，这阻止了模型在推理过程中重新检查、验证或细化视觉证据。虽然工具集成推理提供了一条有前途的路径，但开源VLMs缺乏训练基础设施来学习在多模态医学推理中有效选择、调用和协调工具。我们介绍了MedVistaGym，一种可扩展且交互性强的训练环境，它激励工具集成的视觉推理用于医学图像分析。MedVistaGym使VLMs能够决定何时以及使用哪些工具，定位任务相关的图像区域，并将单个或多个子图像证据整合到交错的多模态推理中，通过统一的可执行接口进行代理训练。使用MedVistaGym，我们训练了MedVistaGym-R1，通过轨迹采样和端到端强化学习将工具使用与代理推理交错进行。在六个医学VQA基准测试中，MedVistaGym-R1-8B在与同样规模的工具增强基线相比，以19.10%至24.21%的优势表现突出，证明了结构化的代理训练——而不是单纯的工具访问——解锁了有效的工具集成推理用于医学图像分析。

英文摘要：

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.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。