TensorX
返回文献探索

Paper · arXiv 2602.02361

SWE-Universe: Scale Real-World Verifiable Environments to Millions

Mouxiang Chen, Lei Zhang, Yunlong Feng, Xuwu Wang, Wenting Zhao, Ruisheng Cao, Jiaxi Yang, Jiawei Chen, Mingze Li, Zeyao Ma, Hao Ge, Zongmeng Zhang, Zeyu Cui, Dayiheng Liu, Jingren Zhou, Jianling Sun, Junyang Lin, Binyuan Hui

61 upvotesFebruary 2, 2026arXiv 预印本
AI 摘要

A scalable framework for constructing real-world software engineering environments from GitHub pull requests using an efficient building agent with self-verification and hacking detection capabilities.

building agentself-verificationin-loop hacking detectionreal-world software engineeringGitHub pull requestscustom-trained modelverifiable environmentsagentic mid-trainingreinforcement learningSWE-Bench Verified

Abstract

We propose SWE-Universe, a scalable and efficient framework for automatically constructing real-world software engineering (SWE) verifiable environments from GitHub pull requests (PRs). To overcome the prevalent challenges of automatic building, such as low production yield, weak verifiers, and prohibitive cost, our framework utilizes a building agent powered by an efficient custom-trained model. This agent employs iterative self-verification and in-loop hacking detection to ensure the reliable generation of high-fidelity, verifiable tasks. Using this method, we scale the number of real-world multilingual SWE environments to a million scale (807,693). We demonstrate the profound value of our environments through large-scale agentic mid-training and reinforcement learning. Finally, we applied this technique to Qwen3-Max-Thinking and achieved a score of 75.3% on SWE-Bench Verified. Our work provides both a critical resource and a robust methodology to advance the next generation of coding agents.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
SWE-Universe: Scale Real-World Verifiable Environments to Millions | TensorX