<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:content="http://purl.org/rss/1.0/modules/content/"><channel><title>漏洞修复 on MessageDaily</title><link>https://inkeast.github.io/MessageDaily/tags/%E6%BC%8F%E6%B4%9E%E4%BF%AE%E5%A4%8D/</link><description>Recent content in 漏洞修复 on MessageDaily</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Sat, 05 Sep 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://inkeast.github.io/MessageDaily/tags/%E6%BC%8F%E6%B4%9E%E4%BF%AE%E5%A4%8D/index.xml" rel="self" type="application/rss+xml"/><item><title>PatchBench 精读：AI 漏洞修复的解决率被高估了 1.83 倍</title><link>https://inkeast.github.io/MessageDaily/posts/2026-09-05-patchbench-vuln-patching-paper-reading/</link><pubDate>Sat, 05 Sep 2026 00:00:00 +0000</pubDate><guid>https://inkeast.github.io/MessageDaily/posts/2026-09-05-patchbench-vuln-patching-paper-reading/</guid><description>马里兰大学的 PatchBench 揭示 AI 漏洞修复评测的两大效度威胁：25% 的 Agent 补丁与历史开发者补丁高度相似（补丁记忆），PoC-only 验证使 11 个 SOTA Agent 的解决率平均虚增 1.83×。其解法是只选 ground-truth 修复在 crash stack 之外的漏洞 + 漏洞移植 + 安全/语义双重验证。本文基于全文阅读拆解 DiffBLEU 记忆检测与验证协议设计。</description></item></channel></rss>