<?xml version="1.0" encoding="utf-8" standalone="yes"?><rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom"><channel><title>AI | Web IR NLP Group @ NUS</title><link>https://knmnyn.github.io/hugo-blox/tag/ai/</link><atom:link href="https://knmnyn.github.io/hugo-blox/tag/ai/index.xml" rel="self" type="application/rss+xml"/><description>AI</description><generator>Hugo Blox Builder (https://hugoblox.com)</generator><language>en-us</language><lastBuildDate>Thu, 30 Apr 2026 11:00:00 +0800</lastBuildDate><image><url>https://knmnyn.github.io/hugo-blox/media/icon_hub1b49484bccf4f586453e7d2a784a896_225489_512x512_fill_lanczos_center_3.png</url><title>AI</title><link>https://knmnyn.github.io/hugo-blox/tag/ai/</link></image><item><title>Lyle Ungar Visits WING to Give Invited NAII Seminar on Evaluating LLMs for Cross-Cultural Coaching and Therapy</title><link>https://knmnyn.github.io/hugo-blox/event/26-04-30-lyle-ungar-naii/</link><pubDate>Thu, 30 Apr 2026 11:00:00 +0800</pubDate><guid>https://knmnyn.github.io/hugo-blox/event/26-04-30-lyle-ungar-naii/</guid><description>&lt;p>Did you miss &lt;strong>Lyle Ungar&lt;/strong>&amp;rsquo;s NUS AI Institute seminar on evaluating large language models for cross-cultural coaching and therapy?&lt;/p>
&lt;p>Watch the recording to learn how frontier LLMs still struggle with culturally appropriate conversation and what can be done about it.&lt;/p>
&lt;p>Lyle Ungar (University of Pennsylvania) explores how to evaluate LLMs on subjective tasks like coaching and therapy across cultures. The talk highlights:&lt;/p>
&lt;ul>
&lt;li>An interpretable framework for comparing linguistic politeness cues across languages&lt;/li>
&lt;li>Evidence that LLMs can homogenize non-Western styles toward stylistic neutrality, a failure missed by standard metrics&lt;/li>
&lt;li>RASTA, a retrieval-based method that preserves culturally appropriate style without hurting content&lt;/li>
&lt;li>The Culturally-Aware Conversations benchmark&lt;/li>
&lt;/ul>
&lt;p>Lyle Ungar is Professor of Computer and Information Science at the University of Pennsylvania, with cross-appointments in Psychology, Bioengineering, Genomics &amp;amp; Computational Biology, and Operations, Information &amp;amp; Decisions. His group builds explainable NLP and socio-emotionally sensitive AI for psychological and medical research.&lt;/p>
&lt;p>This NUS AI Institute seminar was recorded and rebroadcast with explicit permission from Lyle Ungar.&lt;/p>
&lt;p>&lt;strong>Recording:&lt;/strong> &lt;a href="https://www.youtube.com/watch?v=PF2aj1PdwWY" target="_blank" rel="noopener">Watch the seminar&lt;/a>&lt;/p></description></item></channel></rss>