<?xml version="1.0" encoding="utf-8" standalone="yes"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom">
  <channel>
    <title>论文阅读 on BOLIN的个人博客</title>
    <link>https://example.org/papers/</link>
    <description>Recent content in 论文阅读 on BOLIN的个人博客</description>
    <generator>Hugo -- gohugo.io</generator>
    <language>zh-cn</language>
    <copyright>© 2026 BOLIN</copyright>
    <lastBuildDate>Thu, 23 Jul 2026 00:00:00 +0000</lastBuildDate><atom:link href="https://example.org/papers/index.xml" rel="self" type="application/rss+xml" />
    
    <item>
      <title>读懂 DDPM：从加噪过程到噪声预测</title>
      <link>https://example.org/papers/ddpm/</link>
      <pubDate>Thu, 23 Jul 2026 00:00:00 +0000</pubDate>
      
      <guid>https://example.org/papers/ddpm/</guid>
      <description>DDPM 为什么可以从高斯噪声中生成图像？这篇笔记从前向加噪、反向去噪和噪声预测目标三个部分建立直觉。</description>
      
    </item>
    
    <item>
      <title>Restormer：高分辨率图像复原中的高效 Transformer</title>
      <link>https://example.org/papers/restormer/</link>
      <pubDate>Mon, 20 Jul 2026 00:00:00 +0000</pubDate>
      
      <guid>https://example.org/papers/restormer/</guid>
      <description>Restormer 不在空间位置之间构建平方复杂度的注意力，而是在通道维度建模全局关系，并用门控前馈网络保留局部结构。</description>
      
    </item>
    
  </channel>
</rss>
