arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2602.03419cs.SEcs.CL
收藏

SWE-World：在无Docker环境中构建软件工程代理

SWE-World: Building Software Engineering Agents in Docker-Free Environments

Gaoling School of Artificial Intelligence, Renmin University of China（中国人民大学耿丽学校人工智能学院）；BOSS Zhipin, Beijing, China（BOSS智聘，北京，中国）

Shuang Sun, Huatong Song, Lisheng Huang, Jinhao Jiang, Ran Le, Zhihao Lv, Zongchao Chen, Yiwen Hu, Wenyang Luo, Wayne Xin Zhao, Yang Song, Hongteng Xu, Tao Zhang, Ji-Rong Wen

阅读 PDF
arXiv 页面 ↗HTML 原文 ↗

AI总结：

SWE-World提出了一种无Docker环境的框架，利用学习模型预测执行结果，提升软件工程代理性能。

摘要：

近年来，大语言模型（LLMs）的进步使软件工程代理能够处理复杂的代码修改任务。大多数现有方法依赖于容器化环境中的执行反馈，这需要完整的依赖设置和程序及测试的物理执行。虽然有效，但这种范式资源消耗大，难以维护，大大增加了代理训练的难度并限制了可扩展性。我们提出了SWE-World，一个无Docker框架，用学习的替代品取代物理执行环境，用于训练和评估软件工程代理。SWE-World利用基于LLM的模型，这些模型在真实的代理-环境交互数据上进行训练，以预测中间执行结果和最终测试反馈，使代理能够在不与物理容器化环境交互的情况下学习。这种设计保留了标准的代理-环境交互循环，同时在代理优化和评估期间消除了昂贵的环境构建和维护的需要。此外，由于SWE-World可以在不进行真实提交的情况下模拟候选轨迹的最终评估结果，因此它能够从多个测试时间尝试中选择最佳解决方案，从而促进软件工程任务中的有效测试时间扩展（TTS）。在SWE-bench Verified上的实验表明，SWE-World通过无Docker的SFT将Qwen2.5-Coder-32B提升至6.2%到52.0%，通过无Docker的RL提升至55.0%，并通过进一步的TTS提升至68.2%。代码可在https://github.com/RUCAIBox/SWE-World上获得。

英文摘要：

Recent advances in large language models (LLMs) have enabled software engineering agents to tackle complex code modification tasks. Most existing approaches rely on execution feedback from containerized environments, which require dependency-complete setup and physical execution of programs and tests. While effective, this paradigm is resource-intensive and difficult to maintain, substantially complicating agent training and limiting scalability. We propose SWE-World, a Docker-free framework that replaces physical execution environments with a learned surrogate for training and evaluating software engineering agents. SWE-World leverages LLM-based models trained on real agent-environment interaction data to predict intermediate execution outcomes and final test feedback, enabling agents to learn without interacting with physical containerized environments. This design preserves the standard agent-environment interaction loop while eliminating the need for costly environment construction and maintenance during agent optimization and evaluation. Furthermore, because SWE-World can simulate the final evaluation outcomes of candidate trajectories without real submission, it enables selecting the best solution among multiple test-time attempts, thereby facilitating effective test-time scaling (TTS) in software engineering tasks. Experiments on SWE-bench Verified demonstrate that SWE-World raises Qwen2.5-Coder-32B from 6.2\% to 52.0\% via Docker-free SFT, 55.0\% with Docker-free RL, and 68.2\% with further TTS. The code is available at https://github.com/RUCAIBox/SWE-World

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

↑