cs.CLNov 1, 2026

LegoEnv: Self-Evolving Agentic Environments via Modular Recombination

Authors: Zhuosong Jiang, Chengxu Wu, Junqi Liu, Yujie Yi, Siqin Wang, Zhixuan Huang, Jinqian Jiang, Lijun Xiao, 

Organizations: National University of Singapore · Peking University · University of California, Santa Cruz · Shanghai Jiao Tong University · 

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.

Figures & tables

Appendix figures & tables12 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Breaking the Environment Wall: A Unified Framework for Preparing and Evolving Agent-Native Environments

    Sep 24, 2026Yukai Wu, Yuanjing Yang, Le Zhou +8LLM Agent Self-ImprovementLLM Agent Training

  2. EnvSimBench: A Benchmark for Evaluating and Improving LLM-Based Environment Simulation

    May 8, 2026Yi Liu, TingFeng Hui, Wei Zhang +4AI Agent EvaluationAI Agent Benchmarks

  3. EnvFactory: Scaling Tool-Use Agents via Executable Environments Synthesis and Robust RL

    May 18, 2026Minrui Xu, Zilin Wang, Mengyi DENG +12Tool-Augmented Language Model AgentsAgentic RL