<?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>幻觉 on Plbear | 后端架构 · AI工程 · 云原生技术博客</title><link>https://www.plbear.com/tags/%E5%B9%BB%E8%A7%89/</link><description>Recent content in 幻觉 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>Wed, 09 Oct 2024 09:00:00 +0800</lastBuildDate><atom:link href="https://www.plbear.com/tags/%E5%B9%BB%E8%A7%89/index.xml" rel="self" type="application/rss+xml"/><item><title>LLM 幻觉治理：从 Prompt 到 RAG 到评测</title><link>https://www.plbear.com/posts/2024-10-09-llm-hallucination-control/</link><pubDate>Wed, 09 Oct 2024 09:00:00 +0800</pubDate><guid>https://www.plbear.com/posts/2024-10-09-llm-hallucination-control/</guid><description>幻觉是 LLM 的&amp;rsquo;天性&amp;rsquo;，治理是系统工程：Prompt 约束、RAG 兜底、引用强制、评测闭环。把 AI 面试的越级率打到 0% 的幻觉治理体系完整复盘。</description></item></channel></rss>