arXiv cs.AI· Olivier Nourry, Yoshiki Higo·· 12 小时前AI 评分38
Relay：用规则复用与智能体修复扩展历史构建还原
RELAY: Scaling Historical Build Restoration through Rule Reuse and Agentic Repair
AI 导读
Relay 是一个混合框架，先用 LLM 从代表性失败中推导并验证确定性修复规则，复用规则时无需调用模型，规则无法解决的案例再交给逐案例智能体处理。研究基于 14,159 个 OSS-Fuzz 案例，发现 69.7% 已无法完成重建与验证，且多数失败共享少量反复出现的症状。实验显示规则推导可完全自动化，规则优先、智能体兜底的混合方式比单用任一方法还原更多构建，且成本低于纯智能体方案。
正文 · 原文
Abstract:Vulnerability benchmarks and software engineering research depend on rebuilding historical software. Yet builds rely on external resources, such as base images and packages, that change independently of the source code. In this study, we examine this environment decay on 14,159 OSS-Fuzz cases. We find that 69.7% no longer complete reconstruction and verification, and that most failures share a small number of recurring failure symptoms. A recurring failure can therefore be diagnosed once and its repair reused. We present Relay, a hybrid framework that uses an LLM to derive and validate deterministic repair rules from representative failures, reuses them without model calls, and escalates cases unresolved by rules to a per-case agent. We evaluate Relay in human-guided and fully automated rule-derivation campaigns and compare it with a per-case agent approach. We find that rule derivation can be fully automated, and that reusable rules restore as many builds as per-case agents with fewer build executions. However, agents restore cases that the rules cannot. Our results show that a hybrid approach that applies rules first and escalates the remaining cases to an agent restores more builds than either approach alone and at a lower cost than using agents alone.
| Subjects: | Software Engineering (cs.SE); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.27784 [cs.SE] |
| (or arXiv:2610.27784v1 [cs.SE] for this version) | |
| https://doi.org/10.48550/arXiv.2610.27784 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Olivier Nourry [view email]
[v1]
Tue, 20 Oct 2026 07:49:56 UTC (522 KB)
来源：arXiv cs.AI · arxiv.org