MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912628592345088 |
|---|---|
| author | Xu, Ran Zhuang, Yuchen Zhong, Yishan Yu, Yue Wang, Zifeng Tang, Xiangru Wu, Hang Wang, May D. Ruan, Peifeng Yang, Donghan Wang, Tao Xiao, Guanghua Liu, Xin Yang, Carl Xie, Yang Shi, Wenqi |
| author_facet | Xu, Ran Zhuang, Yuchen Zhong, Yishan Yu, Yue Wang, Zifeng Tang, Xiangru Wu, Hang Wang, May D. Ruan, Peifeng Yang, Donghan Wang, Tao Xiao, Guanghua Liu, Xin Yang, Carl Xie, Yang Shi, Wenqi |
| contents | We introduce MedAgentGym, a scalable and interactive training environment designed to enhance coding-based biomedical reasoning capabilities in large language model (LLM) agents. MedAgentGym comprises 72,413 task instances across 129 categories derived from 12 authentic real-world biomedical scenarios. Tasks are encapsulated within executable sandbox environments, each featuring detailed task specifications, interactive feedback mechanisms, verifiable ground truth annotations, and scalable training trajectory generation. Extensive benchmarking of 29 LLMs reveals substantial performance disparities in biomedical data science between commercial and open-source LLMs. Leveraging efficient multi-threaded and multi-turn trajectory sampling in MedAgentGym, Med-Copilot achieves performance gains of +43.02% and +45.28% from offline and online reinforcement learning, respectively, demonstrating MedAgentGym as an effective training ground while establishing itself as a cost-effective, privacy-preserving alternative competitive with proprietary LLMs (gpt-4o). By offering a unified execution environment with a comprehensive benchmark and accessible, extensible training resources, MedAgentGym delivers an integrated platform to develop LLM-based coding assistants for advanced biomedical data science. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_04405 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science Xu, Ran Zhuang, Yuchen Zhong, Yishan Yu, Yue Wang, Zifeng Tang, Xiangru Wu, Hang Wang, May D. Ruan, Peifeng Yang, Donghan Wang, Tao Xiao, Guanghua Liu, Xin Yang, Carl Xie, Yang Shi, Wenqi Computation and Language Artificial Intelligence Machine Learning We introduce MedAgentGym, a scalable and interactive training environment designed to enhance coding-based biomedical reasoning capabilities in large language model (LLM) agents. MedAgentGym comprises 72,413 task instances across 129 categories derived from 12 authentic real-world biomedical scenarios. Tasks are encapsulated within executable sandbox environments, each featuring detailed task specifications, interactive feedback mechanisms, verifiable ground truth annotations, and scalable training trajectory generation. Extensive benchmarking of 29 LLMs reveals substantial performance disparities in biomedical data science between commercial and open-source LLMs. Leveraging efficient multi-threaded and multi-turn trajectory sampling in MedAgentGym, Med-Copilot achieves performance gains of +43.02% and +45.28% from offline and online reinforcement learning, respectively, demonstrating MedAgentGym as an effective training ground while establishing itself as a cost-effective, privacy-preserving alternative competitive with proprietary LLMs (gpt-4o). By offering a unified execution environment with a comprehensive benchmark and accessible, extensible training resources, MedAgentGym delivers an integrated platform to develop LLM-based coding assistants for advanced biomedical data science. |
| title | MedAgentGym: A Scalable Agentic Training Environment for Code-Centric Reasoning in Biomedical Data Science |
| topic | Computation and Language Artificial Intelligence Machine Learning |
| url | https://arxiv.org/abs/2506.04405 |