AgentGym/AgentGym-RL-Data-ID
收藏Hugging Face2025-09-12 更新2025-09-13 收录
下载链接：
https://hf-mirror.com/datasets/AgentGym/AgentGym-RL-Data-ID
下载链接
链接失效反馈官方服务：
资源简介：
AgentGym-RL是一个用于通过多轮强化学习训练大型语言模型（LLM）代理进行长期决策的框架。它支持多种现实世界场景，包括网页导航、深度搜索、数字游戏、具身任务和科学任务，并提供了相应的数据集。
AgentGym-RL is a framework for training large language model (LLM) agents for long-horizon decision-making through multi-turn reinforcement learning. It supports a variety of real-world scenarios including web navigation, deep search, digital games, embodied tasks, and scientific tasks, along with the corresponding dataset.
提供机构：
AgentGym
