MemGovern：透過學習受規範的人類經驗來增強程式碼代理

摘要

儘管自主軟體工程代理正重塑程式設計範式，但目前存在「封閉世界」的侷限性：它們嘗試從零開始或僅依賴本地上下文來修復錯誤，卻忽略了GitHub等平台上可用的龐大人類歷史經驗。由於現實世界問題追蹤數據的非結構化與碎片化特性，存取這些開放世界經驗面臨阻礙。本文提出MemGovern框架，旨在治理原始GitHub數據並將其轉化為代理可操作的經驗記憶。MemGovern透過經驗治理將人類經驗轉換為代理友好的經驗卡片，並引入代理式經驗搜尋策略，實現邏輯驅動的人類專業知識檢索。透過生成13.5萬張經治理的經驗卡片，MemGovern實現了顯著效能提升，在SWE-bench Verified上的問題解決率提高4.65%。作為外掛式解決方案，MemGovern為代理友好型記憶基礎設施提供了創新實踐。

English

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.

MemGovern：透過學習受規範的人類經驗來增強程式碼代理

MemGovern: Enhancing Code Agents through Learning from Governed Human Experiences

摘要

Support