MedAgentGYM, a training environment for coding-based medical reasoning in LLMs, enhances performance through supervised fine-tuning and reinforcement learning, providing a benchmark and expandable resource.
We introduce MedAgentGYM, the first publicly available training environment
designed to enhance coding-based medical reasoning capabilities in large
language model (LLM) agents. MedAgentGYM comprises 72,413 task instances across
129 categories derived from authentic real-world biomedical scenarios. Tasks
are encapsulated within executable coding environments, each featuring detailed
task descriptions, interactive feedback mechanisms, verifiable ground-truth
annotations, and scalable training trajectory generation. Extensive
benchmarking of over 30 LLMs reveals a notable performance disparity between
commercial API-based models and open-source counterparts. Leveraging
MedAgentGYM, Med-Copilot-7B achieves substantial performance gains through
supervised fine-tuning (+36.44%) and continued reinforcement learning
(+42.47%), emerging as an affordable and privacy-preserving alternative
competitive with gpt-4o. By offering both a comprehensive benchmark and
accessible, expandable training resources within unified execution
environments, MedAgentGYM delivers an integrated platform to develop LLM-based
coding assistants for advanced biomedical research and practice.