AgentGym: Evolving Large Language Model-based Agents across Diverse Environments

Source: arXiv Authors: Zhiheng Xi, Yiwen Ding, Wenxiang Chen, Boyang Hong, Honglin Guo, Junzhe Wang, Dingwen Yang, Chenyang Liao, et al. Date: 2024-06-06 Primary category: cs.AI All categories: cs.AI, cs.CL

Abstract

AgentGym is a broad multi-environment framework explicitly aimed at agent self-evolution rather than single-environment specialization. It is one of the papers most directly aligned with the question of RL-gym-like substrates for general harness improvement.