<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>xieydd on VONNG</title><link>https://blog.vonng.com/authors/xieydd/</link><description>Recent content in xieydd on VONNG</description><generator>Hugo</generator><language>zh-CN</language><lastBuildDate>Mon, 23 Mar 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://blog.vonng.com/authors/xieydd/index.xml" rel="self" type="application/rss+xml"/><item><title>从 KV Cache 到 AI 内存系统：大模型推理架构的演进</title><link>https://blog.vonng.com/ai/kv-cache-memory-system/</link><pubDate>Mon, 23 Mar 2026 00:00:00 +0000</pubDate><guid>https://blog.vonng.com/ai/kv-cache-memory-system/</guid><description>摘要 这篇文章想回答一个看似分散、其实高度统一的问题：为什么这两年围绕大模型推理的系统创新，越来越不像是在“优化一个神经网络”，反而像是在“设计一个内存系统”？</description></item></channel></rss>