arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2601.06789cs.SEcs.AI
收藏

MemGovern:通过学习受控的人类经验增强代码代理

MemGovern: Enhancing Code Agents through Learning from Governed Human Experiences

UCAS(中国科学院大学);NUS(新加坡国立大学);ECNU(华东师范大学);FDU(福建师范大学);XDU(西安电子科技大学);UBC(不列颠哥伦比亚大学);HKUST(GZ)(香港科技大学);PKU(北京大学);RUC(中国人民大学);QuantaAlpha(量子阿尔法)

Qihao Wang, Ziming Cheng, Shuo Zhang, Fan Liu, Rui Xu, Heng Lian, Kunyi Wang, Xiaoming Yu, Jianghao Yin, Sen Hu, Yue Hu, Shaolei Zhang, Yanbing Liu, Ronghao Chen, Huacan Wang

AI总结:

MemGovern通过学习受控的人类经验,提升代码代理的修复效率和性能。

摘要:

尽管自主软件工程(SWE)代理正在重塑编程范式,但它们目前受到

英文摘要:

While autonomous software engineering (SWE) agents are reshaping programming paradigms, they currently suffer from a "closed-world" limitation: they attempt to fix bugs from scratch or solely using local context, ignoring the immense historical human experience available on platforms like GitHub. Accessing this open-world experience is hindered by the unstructured and fragmented nature of real-world issue-tracking data. In this paper, we introduce MemGovern, a framework designed to govern and transform raw GitHub data into actionable experiential memory for agents. MemGovern employs experience governance to convert human experience into agent-friendly experience cards and introduces an agentic experience search strategy that enables logic-driven retrieval of human expertise. By producing 135K governed experience cards, MemGovern achieves a significant performance boost, improving resolution rates on the SWE-bench Verified by 4.65%. As a plug-in approach, MemGovern provides a solution for agent-friendly memory infrastructure.

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。

↑