<?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>Evals on My Learning Notes</title>
    <link>https://learning-notes-dz2.pages.dev/tags/evals/</link>
    <description>Recent content in Evals on My Learning Notes</description>
    <generator>Hugo -- 0.124.0</generator>
    <language>en</language>
    <lastBuildDate>Sun, 11 Oct 2026 13:39:52 +0700</lastBuildDate>
    <atom:link href="https://learning-notes-dz2.pages.dev/tags/evals/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Before You Trust an Eval Number, Find the Useless System That Would Score Well on It</title>
      <link>https://learning-notes-dz2.pages.dev/posts/2026-10-17/</link>
      <pubDate>Sat, 10 Oct 2026 08:43:00 +0700</pubDate>
      <guid>https://learning-notes-dz2.pages.dev/posts/2026-10-17/</guid>
      <description>A judge that always picks position A is perfectly consistent, and an agent that abstains on everything never makes a confident mistake. Every evaluation number needs a degenerate baseline before it means anything.</description>
    </item>
  </channel>
</rss>
