<?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>Mutation-Testing on 兜忆轩 | Cedric's Compiler Notes</title><link>https://douyixuan.github.io/tags/mutation-testing/</link><description>Recent content in Mutation-Testing on 兜忆轩 | Cedric's Compiler Notes</description><image><title>兜忆轩 | Cedric's Compiler Notes</title><url>https://douyixuan.github.io/images/papermod-cover.png</url><link>https://douyixuan.github.io/images/papermod-cover.png</link></image><generator>Hugo -- 0.146.0</generator><language>en-us</language><lastBuildDate>Thu, 15 Oct 2026 12:39:00 +0800</lastBuildDate><atom:link href="https://douyixuan.github.io/tags/mutation-testing/index.xml" rel="self" type="application/rss+xml"/><item><title>030 · 只有 0.48% 通过，是模型失败还是评测失效？</title><link>https://douyixuan.github.io/posts/ai-silicon-papers/2026-10-15/</link><pubDate>Thu, 15 Oct 2026 12:39:00 +0800</pubDate><guid>https://douyixuan.github.io/posts/ai-silicon-papers/2026-10-15/</guid><description>SecTB-RTL 的 1,860 次调用只有 9 次通过执行端语义校验，但生成约定遗漏了执行端的复位规则。作者因此保留失败记录，却不报告提示词效果：完整分母也救不了失配的测量工具。</description></item></channel></rss>