<?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>推理加速 on Ge Zhang · 技术笔记</title>
    <link>https://blog-shf.pages.dev/tags/%E6%8E%A8%E7%90%86%E5%8A%A0%E9%80%9F/</link>
    <description>Recent content in 推理加速 on Ge Zhang · 技术笔记</description>
    <generator>Hugo</generator>
    <language>zh</language>
    <lastBuildDate>Tue, 14 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://blog-shf.pages.dev/tags/%E6%8E%A8%E7%90%86%E5%8A%A0%E9%80%9F/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Token Recycling：把被丢弃的候选 Token 变成下一轮草稿</title>
      <link>https://blog-shf.pages.dev/papers/token-recycling/</link>
      <pubDate>Tue, 14 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://blog-shf.pages.dev/papers/token-recycling/</guid>
      <description>无需训练草稿模型，Token Recycling 用一个不到 2 MiB 的邻接矩阵回收目标模型已经算出的 top-k 候选，并通过草稿树与 Tree Attention 加速解码。</description>
    </item>
  </channel>
</rss>
