<?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>📊 评估体系 on Elon&#39;s AD Insight</title>
    <link>https://auto-driving-blog.vercel.app/tags/-%E8%AF%84%E4%BC%B0%E4%BD%93%E7%B3%BB/</link>
    <description>Recent content in 📊 评估体系 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>Sun, 19 Jul 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://auto-driving-blog.vercel.app/tags/-%E8%AF%84%E4%BC%B0%E4%BD%93%E7%B3%BB/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>模型评估体系与回归检测：守护版本质量</title>
      <link>https://auto-driving-blog.vercel.app/posts/knowledge/%E6%A8%A1%E5%9E%8B%E8%AF%84%E4%BC%B0%E4%BD%93%E7%B3%BB%E4%B8%8E%E5%9B%9E%E5%BD%92%E6%A3%80%E6%B5%8B/</link>
      <pubDate>Sun, 19 Jul 2026 00:00:00 +0000</pubDate>
      <guid>https://auto-driving-blog.vercel.app/posts/knowledge/%E6%A8%A1%E5%9E%8B%E8%AF%84%E4%BC%B0%E4%BD%93%E7%B3%BB%E4%B8%8E%E5%9B%9E%E5%BD%92%E6%A3%80%E6%B5%8B/</guid>
      <description>自动驾驶模型需要频繁迭代，但如何确保新版本&amp;rsquo;不比旧版本差&amp;rsquo;是一个被严重低估的工程难题。本文讲解模型回归检测的全套实践：离线评估Pipeline搭建（开环感知指标/闭环规划指标/计算延迟）、回归测试集的设计原则（场景覆盖/难度层级/对抗样本）、指标一致性分析（confidence interval/statistical significance/effect size）、以及CI/CD流水线中的自动化门禁机制（regression gate+人工review）。</description>
    </item>
  </channel>
</rss>
