BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Large Language Models\, Model Collapse\, and the Conservation of I
 nformation - George Montanez
DTSTART:20260217T150000Z
DTEND:20260217T160000Z
UID:TALK243508@talks.cam.ac.uk
CONTACT:Pietro Lio
DESCRIPTION:Do Large Language Models (LLMs) think and reason? Are they per
 petual information machines\, producing endless coherent and correct text 
 from finite training data? We explore how LLMs work and whether they produ
 ce rational thought and endless information. We show how theoretical consi
 derations and experimental results from philosophy\, statistics\, informat
 ion theory\, and machine learning argue against the thesis that LLMs are r
 ational\, information-generating entities.
LOCATION:Computer Laboratory\, William Gates Building\, Room LT1
END:VEVENT
END:VCALENDAR