<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI Evaluations | Lawrence Chan</title><link>https://chanlawrence.me/tag/ai-evaluations/</link><atom:link href="https://chanlawrence.me/tag/ai-evaluations/index.xml" rel="self" type="application/rss+xml"/><description>AI Evaluations</description><generator>Wowchemy (https://wowchemy.com)</generator><language>en-us</language><lastBuildDate>Wed, 19 Mar 2025 00:00:00 +0000</lastBuildDate><image><url>https://chanlawrence.me/media/sharing.jpeg</url><title>AI Evaluations</title><link>https://chanlawrence.me/tag/ai-evaluations/</link></image><item><title>Measuring AI Ability to Complete Long Software Tasks</title><link>https://chanlawrence.me/publication/kwa-2025-measuring/</link><pubDate>Wed, 19 Mar 2025 00:00:00 +0000</pubDate><guid>https://chanlawrence.me/publication/kwa-2025-measuring/</guid><description/></item><item><title>HCAST: Human-Calibrated Autonomy Software Tasks</title><link>https://chanlawrence.me/publication/rein-2025-hcast/</link><pubDate>Tue, 18 Mar 2025 00:00:00 +0000</pubDate><guid>https://chanlawrence.me/publication/rein-2025-hcast/</guid><description/></item></channel></rss>