arXiv cs.CL· Zhuosong Jiang, Chengxu Wu, Junqi Liu, Yujie Yi, Siqin Wang, Zhixuan Huang, Jinqian Jiang, Lijun Xiao, Jianxun Li, Yongyang Pan, Jieran Wang, Shiqi Shen, Fei Liu, Xinyu Lin·· 11 小时前AI 评分45
LegoEnv：通过模块化重组实现自进化智能体环境
LegoEnv: Self-Evolving Agentic Environments via Modular Recombination
AI 导读
LegoEnv 提出"环境原子"新范式，用可复用的 intent、rule、data、server 原子组合环境，并在环境反馈不可靠时进化原子，已构建覆盖 223 个应用的 5,465 个经验证环境。
正文 · 原文
Authors:Zhuosong Jiang, Chengxu Wu, Junqi Liu, Yujie Yi, Siqin Wang, Zhixuan Huang, Jinqian Jiang, Lijun Xiao, Jianxun Li, Yongyang Pan, Jieran Wang, Shiqi Shen, Fei Liu, Xinyu Lin
Abstract:LLM agents learn by interacting with environments, yet existing environments are designed for specific tasks and fail to generalize to new ones. Recent methods automate environment construction with LLMs. However, they generate each environment from scratch, neither reusing parts shared across environments nor carrying a fix from one environment to the rest. To address this, we propose a new paradigm of evolving environment atoms, which composes environments from reusable atoms (intent, rule, data and server) and evolves the atoms when an environment fails to give reliable feedback (e.g., an application lacks a tool the task requires). To instantiate this paradigm, we introduce LegoEnv, which builds 5,465 verified environments over 223 applications, and one round of evolution raises the share of solvable environments from 71.6% to 88.3%. On AutomationBench, SFT on these environments lifts Qwen3.6-27B from 16.0% to 32.0%, and RL with their verifiers lifts Qwen3.8-27B from 37.6% to 40.3%. Further analyses show steady gains with corpus size, transfer to unseen applications, and failures concentrated in a few shared root causes.
| Subjects: | Computation and Language (cs.CL); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.01834 [cs.CL] |
| (or arXiv:2611.01834v1 [cs.CL] for this version) | |
| https://doi.org/10.48550/arXiv.2611.01834 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zhuosong Jiang [view email]
[v1]
Sun, 1 Nov 2026 12:07:14 UTC (503 KB)
来源：arXiv cs.CL · arxiv.org