<?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>Rag on Dingfan X.</title><link>https://dingfanx.com/tags/rag/</link><description>Recent content in Rag on Dingfan X.</description><generator>Hugo</generator><language>zh-cn</language><copyright>2026 Dingfan. 本站内容采用 CC BY-NC-SA 4.0 协议</copyright><lastBuildDate>Tue, 05 May 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://dingfanx.com/tags/rag/index.xml" rel="self" type="application/rss+xml"/><item><title>Chunking Strategies for RAG</title><link>https://dingfanx.com/tech/chunking-strategies-for-rag/</link><pubDate>Tue, 05 May 2026 00:00:00 +0000</pubDate><guid>https://dingfanx.com/tech/chunking-strategies-for-rag/</guid><description>基于 Pinecone Chunking Strategies 的精读笔记。重点整理了 chunk size 的双重影响、embedding 容量瓶颈与 lost-in-the-middle 的本质区别、RAG vs long context 的工程权衡、以及切分与增强的正交矩阵思维。</description></item></channel></rss>