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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Xi, Zhiheng, Ding, Yiwen, Chen, Wenxiang, Hong, Boyang, Guo, Honglin, Wang, Junzhe, Yang, Dingwen, Liao, Chenyang, Guo, Xin, He, Wei, Gao, Songyang, Chen, Lu, Zheng, Rui, Zou, Yicheng, Gui, Tao, Zhang, Qi, Qiu, Xipeng, Huang, Xuanjing, Wu, Zuxuan, Jiang, Yu-Gang
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916277980758016
author Xi, Zhiheng
Ding, Yiwen
Chen, Wenxiang
Hong, Boyang
Guo, Honglin
Wang, Junzhe
Yang, Dingwen
Liao, Chenyang
Guo, Xin
He, Wei
Gao, Songyang
Chen, Lu
Zheng, Rui
Zou, Yicheng
Gui, Tao
Zhang, Qi
Qiu, Xipeng
Huang, Xuanjing
Wu, Zuxuan
Jiang, Yu-Gang
author_facet Xi, Zhiheng
Ding, Yiwen
Chen, Wenxiang
Hong, Boyang
Guo, Honglin
Wang, Junzhe
Yang, Dingwen
Liao, Chenyang
Guo, Xin
He, Wei
Gao, Songyang
Chen, Lu
Zheng, Rui
Zou, Yicheng
Gui, Tao
Zhang, Qi
Qiu, Xipeng
Huang, Xuanjing
Wu, Zuxuan
Jiang, Yu-Gang
contents Building generalist agents that can handle diverse tasks and evolve themselves across different environments is a long-term goal in the AI community. Large language models (LLMs) are considered a promising foundation to build such agents due to their generalized capabilities. Current approaches either have LLM-based agents imitate expert-provided trajectories step-by-step, requiring human supervision, which is hard to scale and limits environmental exploration; or they let agents explore and learn in isolated environments, resulting in specialist agents with limited generalization. In this paper, we take the first step towards building generally-capable LLM-based agents with self-evolution ability. We identify a trinity of ingredients: 1) diverse environments for agent exploration and learning, 2) a trajectory set to equip agents with basic capabilities and prior knowledge, and 3) an effective and scalable evolution method. We propose AgentGym, a new framework featuring a variety of environments and tasks for broad, real-time, uni-format, and concurrent agent exploration. AgentGym also includes a database with expanded instructions, a benchmark suite, and high-quality trajectories across environments. Next, we propose a novel method, AgentEvol, to investigate the potential of agent self-evolution beyond previously seen data across tasks and environments. Experimental results show that the evolved agents can achieve results comparable to SOTA models. We release the AgentGym suite, including the platform, dataset, benchmark, checkpoints, and algorithm implementations. The AgentGym suite is available on https://github.com/WooooDyy/AgentGym.
format Preprint
id arxiv_https___arxiv_org_abs_2406_04151
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
Xi, Zhiheng
Ding, Yiwen
Chen, Wenxiang
Hong, Boyang
Guo, Honglin
Wang, Junzhe
Yang, Dingwen
Liao, Chenyang
Guo, Xin
He, Wei
Gao, Songyang
Chen, Lu
Zheng, Rui
Zou, Yicheng
Gui, Tao
Zhang, Qi
Qiu, Xipeng
Huang, Xuanjing
Wu, Zuxuan
Jiang, Yu-Gang
Artificial Intelligence
Computation and Language
Building generalist agents that can handle diverse tasks and evolve themselves across different environments is a long-term goal in the AI community. Large language models (LLMs) are considered a promising foundation to build such agents due to their generalized capabilities. Current approaches either have LLM-based agents imitate expert-provided trajectories step-by-step, requiring human supervision, which is hard to scale and limits environmental exploration; or they let agents explore and learn in isolated environments, resulting in specialist agents with limited generalization. In this paper, we take the first step towards building generally-capable LLM-based agents with self-evolution ability. We identify a trinity of ingredients: 1) diverse environments for agent exploration and learning, 2) a trajectory set to equip agents with basic capabilities and prior knowledge, and 3) an effective and scalable evolution method. We propose AgentGym, a new framework featuring a variety of environments and tasks for broad, real-time, uni-format, and concurrent agent exploration. AgentGym also includes a database with expanded instructions, a benchmark suite, and high-quality trajectories across environments. Next, we propose a novel method, AgentEvol, to investigate the potential of agent self-evolution beyond previously seen data across tasks and environments. Experimental results show that the evolved agents can achieve results comparable to SOTA models. We release the AgentGym suite, including the platform, dataset, benchmark, checkpoints, and algorithm implementations. The AgentGym suite is available on https://github.com/WooooDyy/AgentGym.
title AgentGym: Evolving Large Language Model-based Agents across Diverse Environments
topic Artificial Intelligence
Computation and Language
url https://arxiv.org/abs/2406.04151