BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Replicating and auditing black-box Language Models. - Tatsunori Ha
 shimoto
DTSTART:20240125T160000Z
DTEND:20240125T170000Z
UID:TALK211336@talks.cam.ac.uk
CONTACT:Panagiotis Fytas
DESCRIPTION:Advances in large language models have brought about exciting 
 advancements in capabilities\, but the commercialization of this technolog
 y has led to an increasing loss of transparency. State-of-the-art language
  models effectively operate as black boxes\, with many things unknown abou
 t their training algorithms\, data annotators\, and pretraining data. I wi
 ll cover a trio of recent works from my group that attempt to help us unde
 rstand each of these components by replicating the RLHF training process (
 AlpacaFarm)\, probing LMs to identify whose opinions are being reflected i
 n pretraining and RLHF data (OpinionQA)\, and providing provable guarantee
 s of test set contamination in black-box language models.
LOCATION:https://cam-ac-uk.zoom.us/j/97599459216?pwd=QTRsOWZCOXRTREVnbTJBd
 XVpOXFvdz09
END:VEVENT
END:VCALENDAR