cs.CLNov 16, 2026

From Repositories to Training Environments: Scaling Long-Horizon Coding Agents with Environment Evolution Loop

Authors: Yang Tian, Tianyu Zheng, Xiyuan Yang, Yaxin Du, Chengran Yang, Bo Zhao, Jian Yang, Bryan Dai

Organizations: School of AI, Shanghai Jiao Tong University · IQuest Research · Beihang University

Abstract

Training coding agents for long-horizon tasks requires environments with verifiable feedback. Existing pipelines often follow a linear process that ends with validation, establishing usability but providing little guidance on improving training value. Recursive self-improvement (RSI) often requires model rollouts or model training followed by downstream evaluation, with costs and feedback delays amplified on long-horizon tasks. How can we use timely, reliable proxy signals to improve training value and synthesize environments at scale? We present RepoFactory, an automated pipeline that converts real repositories into verifiable environments where agents build complete projects from scratch given a language specification. Its Environment Evolution Loop optimizes a multiplicative objective combining validity, structural coupling, and implementation scope through proxy feedback. RepoFactory produces 10,159 environments across 17 software domains. Training Qwen3.5-35B-A3B on these environments yields RepoHorizen-35B-A3B, which achieves gains of 22.6 percentage points on NL2Repo-Bench and 8.5 points on average across five benchmarks covering long-horizon coding and agentic tasks in real-world production settings, compared with 17.5 and 6.4 points without environment evolution. These results highlight the training value of RepoFactory environments and the additional gains from the Environment Evolution Loop.

Figures & tables

Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. ProgramGym: Scaling Training Environments for Long-Horizon Program Reproduction Agents

    Oct 9, 2026Chengyi Yang, Renxu Bao, Jue Hou +10AI Coding AgentsLLM Agent Training

  2. CodeMidas: Scaling Agentic Coding RL Environments from Code Itself

    Sep 18, 2026Bowen Ye, Lei Li, Shicheng Li +16LLM Agent TrainingRL for Code Generation

  3. SPROUT: Scaling Agentic Reinforcement Learning for Long-Horizon Software Engineering

    Oct 9, 2026Mengyang Liu, Weiming Wang, Haixin Nan +6Agentic RLRL for Code Generation