<?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>评测 on Plbear | 后端架构 · AI工程 · 云原生技术博客</title><link>https://www.plbear.com/tags/%E8%AF%84%E6%B5%8B/</link><description>Recent content in 评测 on Plbear | 后端架构 · AI工程 · 云原生技术博客</description><generator>Hugo</generator><language>zh-cn</language><lastBuildDate>Tue, 02 Jun 2026 09:00:00 +0800</lastBuildDate><atom:link href="https://www.plbear.com/tags/%E8%AF%84%E6%B5%8B/index.xml" rel="self" type="application/rss+xml"/><item><title>RAG 评测：离线指标与线上效果的鸿沟</title><link>https://www.plbear.com/posts/2026-06-02-rag-evaluation/</link><pubDate>Tue, 02 Jun 2026 09:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2026-06-02-rag-evaluation/</guid><description>离线评测 recall@5 到 90%，上线后用户还是说&amp;rsquo;答非所问&amp;rsquo;？检索质量、生成质量、端到端三个层面，RAG 评测的真实方法与鸿沟分析。</description></item></channel></rss>