titleFrom Repositories to Training Environments: Scaling Long-Horizon Coding Agents with Environment Evolution Loop
authorsYang Tian, Tianyu Zheng, Xiyuan Yang, Yaxin Du, Chengran Yang, Bo Zhao, Jian Yang, Bryan Dai
date2026-11-17
venuearXiv (preprint, not peer reviewed)
idarXiv:2611.21152
affiliationIQuest
upvotes12 (Hugging Face Daily Papers, 2026-11-19)
linksabs pdf hf code
cite
relatedIQuest-Q1: Advancing Foundation Capabilities for Agentic CLI Systems
Loop the Loopies!
TMAS: Scaling Test-Time Compute via Multi-Agent Synergy
WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training
CyberFactory: Scaling Cyber Security Capabilities with Instances from the Wild
abstractTraining 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.

bibtex

@misc{tian2026from,
  title={From Repositories to Training Environments: Scaling Long-Horizon Coding Agents with Environment Evolution Loop},
  author={Yang Tian and Tianyu Zheng and Xiyuan Yang and Yaxin Du and Chengran Yang and Bo Zhao and Jian Yang and Bryan Dai},
  year={2026},
  eprint={2611.21152},
  archivePrefix={arXiv},
  url={https://arxiv.org/abs/2611.21152}
}