<?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>SparseDriveV2 on Elon&#39;s AD Insight</title>
    <link>https://auto-driving-blog.vercel.app/tags/sparsedrivev2/</link>
    <description>Recent content in SparseDriveV2 on Elon&#39;s AD Insight</description>
    <image>
      <title>Elon&#39;s AD Insight</title>
      <url>https://auto-driving-blog.vercel.app/images/share.png</url>
      <link>https://auto-driving-blog.vercel.app/images/share.png</link>
    </image>
    <generator>Hugo</generator>
    <language>zh-cn</language>
    <lastBuildDate>Mon, 20 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://auto-driving-blog.vercel.app/tags/sparsedrivev2/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>代码讲解：SparseDriveV2 因子化轨迹词汇表与两级评分的完整闭环</title>
      <link>https://auto-driving-blog.vercel.app/posts/code/sparsedrivev2%E4%BB%A3%E7%A0%81%E8%AE%B2%E8%A7%A3/</link>
      <pubDate>Mon, 20 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://auto-driving-blog.vercel.app/posts/code/sparsedrivev2%E4%BB%A3%E7%A0%81%E8%AE%B2%E8%A7%A3/</guid>
      <description>逐行拆解 swc-17/SparseDriveV2 的真实源码：从架构图与全局数据流出发，用最通俗的大白话讲清几何路径乘以速度剖面如何因子化组合成 26 万条轨迹词汇表、两级评分怎么把候选从 26 万粗筛到 200 条再精排、PDM score 蒸馏损失如何用排行榜判分器当老师。一篇把项目挂载方式、数据流、前向传播、训练梯度、推理选轨迹、个人思考全部讲透、面向零基础读者的工程向代码讲解。</description>
    </item>
  </channel>
</rss>
