<?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>Physical Reasoning on Cognaptus</title>
    <link>https://cognaptus.com/tags/physical-reasoning/</link>
    <description>Recent content in Physical Reasoning on Cognaptus</description>
    <generator>Hugo -- 0.145.0</generator>
    <language>en-us</language>
    <lastBuildDate>Thu, 16 Apr 2026 00:00:00 +0000</lastBuildDate>
    <atom:link href="https://cognaptus.com/tags/physical-reasoning/index.xml" rel="self" type="application/rss+xml" />
    <item>
      <title>Rewarding Bad Physics Habits: What VLMs Learn When You Pay Them to Reason</title>
      <link>https://cognaptus.com/blog/2026-04-16-rewarding-bad-physics-habits-what-vlms-learn-when-you-pay-them-to-reason/</link>
      <pubDate>Thu, 16 Apr 2026 00:00:00 +0000</pubDate>
      <guid>https://cognaptus.com/blog/2026-04-16-rewarding-bad-physics-habits-what-vlms-learn-when-you-pay-them-to-reason/</guid>
      <description>A reward-ablation study on VLM physics reasoning shows why accuracy, reasoning discipline, and visual grounding must be treated as different deployment objectives, not one magical intelligence switch.</description>
    </item>
  </channel>
</rss>
