<?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>📚 综述 on Elon&#39;s AD Insight</title>
    <link>https://auto-driving-blog.vercel.app/tags/-%E7%BB%BC%E8%BF%B0/</link>
    <description>Recent content in 📚 综述 on Elon&#39;s AD Insight</description>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Sun, 19 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://auto-driving-blog.vercel.app/tags/-%E7%BB%BC%E8%BF%B0/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>论文精读｜A Survey of World Models for Autonomous Driving：自动驾驶世界模型全景综述</title>
      <link>https://auto-driving-blog.vercel.app/posts/paper-reading/%E8%AE%BA%E6%96%87%E7%B2%BE%E8%AF%BB-2501-11260/</link>
      <pubDate>Sun, 19 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://auto-driving-blog.vercel.app/posts/paper-reading/%E8%AE%BA%E6%96%87%E7%B2%BE%E8%AF%BB-2501-11260/</guid>
      <description>这篇持续更新的综述为自动驾驶世界模型构建了一套三层分类法：Tier I 生成未来物理世界（Image/BEV/OG/PC 四类生成范式）、Tier II 智能体行为规划（规则与学习双轨）、Tier III 预测与规划交互。它系统梳理了自监督学习、多模态预训练、生成式数据增强等训练范式，并配套发布 WMAD 论文清单与基准。是当前入门自动驾驶世界模型最系统的地图。</description>
    </item>
    <item>
      <title>论文精读｜VLA for Autonomous Driving：自动驾驶视觉-语言-动作模型的过去、现在与未来</title>
      <link>https://auto-driving-blog.vercel.app/posts/paper-reading/%E8%AE%BA%E6%96%87%E7%B2%BE%E8%AF%BB-2512-16760/</link>
      <pubDate>Sun, 19 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://auto-driving-blog.vercel.app/posts/paper-reading/%E8%AE%BA%E6%96%87%E7%B2%BE%E8%AF%BB-2512-16760/</guid>
      <description>这份 47 页的综述首次为自动驾驶 VLA 建立了系统形式化与分类法，从 VA（视觉-动作）到 VLA（视觉-语言-动作）理清演进线。它提出端到端 VLA 与双系统 VLA 两大范式分类，并统一梳理了动作空间四分法、数据集基准与评测协议。是当前自动驾驶 VLA 领域最权威的架构脉络地图和未来挑战指南。</description>
    </item>
  </channel>
</rss>
