<?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>🏭 乐博尼克 on Elon&#39;s AD Insight</title>
    <link>https://auto-driving-blog.vercel.app/tags/-%E4%B9%90%E5%8D%9A%E5%B0%BC%E5%85%8B/</link>
    <description>Recent content in 🏭 乐博尼克 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/-%E4%B9%90%E5%8D%9A%E5%B0%BC%E5%85%8B/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>论文精读｜LingBot-VLA：务实的 VLA 基础模型</title>
      <link>https://auto-driving-blog.vercel.app/posts/paper-reading/%E8%AE%BA%E6%96%87%E7%B2%BE%E8%AF%BB-2601-18692/</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-2601-18692/</guid>
      <description>LingBot-VLA 以&amp;rsquo;务实&amp;rsquo;为核心设计哲学，用 9 种双臂构型、约 2 万小时真机数据训出一个跨平台 VLA 基础模型。它通过跨构型预训练学出与具体硬件解耦的动作策略，辅以开源高效代码库（8 卡 261 样本/秒）和 GM-100 基准。在 4 个平台 100 类任务上展现强泛化能力，将 VLA 从论文模型推向产线工具。</description>
    </item>
  </channel>
</rss>
