arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2510.03217cs.SEcs.AI
收藏

回避与验证:一种双LLM策略用于减少代理程序修复中的噪声

Abstain and Validate: A Dual-LLM Policy for Reducing Noise in Agentic Program Repair

Google, USA(谷歌公司);Meta, USA(Meta公司)

José Cambronero, Michele Tufano, Sherry Shi, Renyao Wei, Grant Uy, Runxiang Cheng, Chin-Jung Liu, Shiying Pan, Satish Chandra, Pat Rondon

AI总结:

本文提出双LLM策略,通过bug回避和补丁验证减少代理程序修复中的噪声,提升修复成功率和可靠性。

摘要:

代理自动程序修复(APR)正在越来越多地解决工业中复杂的、仓库级的bug,但最终这些补丁仍需要由人类审查才能提交,以确保它们解决了bug。展示可能不被接受的补丁会导致大量噪声,浪费宝贵的发展者时间并削弱对自动代码更改的信任。我们引入了两种互补的基于LLM的策略来减少这种噪声:bug回避和补丁验证策略。bug回避排除代理APR系统可能无法修复的bug。补丁验证拒绝可能不是给定bug良好修复的补丁。我们在Google代码库的三个bug集上评估了这两种策略,以及由内部代理APR系统生成的候选补丁。在一组174个人报告的bug上,移除被我们策略拒绝的bug和补丁可以将成功率达到13个百分点和15个百分点,组合起来可提高39个百分点。在空指针异常和sanitizer报告的bug中,补丁验证也提高了平均单样本成功率。这种双策略方法为代理APR系统的可靠、大规模工业部署提供了可行的途径。

英文摘要:

Agentic Automated Program Repair (APR) is increasingly tackling complex, repository-level bugs in industry, but ultimately these patches still need to be reviewed by a human before committing them to ensure they address the bug. Showing patches unlikely to be accepted can lead to substantial noise, wasting valuable developer time and eroding trust in automated code changes. We introduce two complementary LLM-based policies to reduce such noise: bug abstention and patch validation policies. Bug abstention excludes bugs that the agentic APR system is unlikely to fix. Patch validation rejects patches that are unlikely to be a good fix for the given bug. We evaluate both policies on three sets of bugs from Google's codebase, and their candidate patches generated by an internal agentic APR system. On a set of 174 human-reported bugs, removing bugs and patches rejected by our policies can raise success rates by up to 13 percentage points and 15 percentage points, respectively, and by up to 39 percentage points in combination. On null pointer exceptions and sanitizer-reported bugs with machine-generated bug reports, patch validation also improves average single-sample success rates. This two-policy approach provides a practical path to the reliable, industrial-scale deployment of agentic APR systems.

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。

↑