Friday, November 13, 2026
11:00 AM – 12:00 PM CST
Teaching Agents That Use Language to Use Language (UT Forum for AI)
About this event
A UT Austin Forum for Artificial Intelligence research talk by UC Berkeley postdoc David Chan on giving multimodal conversational agents a sense of pragmatics, to reduce hallucination and misread requests. In person in GDC 1.304 or on Zoom.
Multimodal conversational agents and pragmatics
The UT Austin Forum for Artificial Intelligence hosts David Chan, a postdoctoral researcher at UC Berkeley and incoming assistant professor at UC Irvine, for a talk titled "Teaching Agents That Use Language to Use Language."
Chan argues that today's chat agents sound fluent yet often fail to notice when a request is impossible, unsupported by the available context, or misunderstood. He ties problems such as hallucination, alignment and adaptation in multimodal LLMs to a missing sense of pragmatics, the implicit context behind a request, and presents work toward conversational agents that can see, listen, act and speak in context. His research covers multimodal conversational AI that combines vision, audio and language.
Practical details
- When: Friday, November 13, 2026, 11:00 a.m. (the page gives no end time, so about one hour is assumed).
- Where: GDC 1.304 in the Gates Dell Complex on the UT Austin campus, or join by Zoom (link on the Forum for AI page).
- Cost: The page lists no fee. Confirm access details on the Forum for AI page.
The Forum for AI is a recurring academic talk series, so expect a research-seminar format.
Organized by
The long-running AI colloquium of the Department of Computer Science at The University of Texas at Austin. FAI brings visiting and local researchers to campus through the academic year for hour-long technical talks spanning machine learning, robotics, language, and human-AI interaction. Talks are free, open to the public, held in the Gates Computer Science Complex, and simulcast on Zoom.