Vered Shwartz (UBC Computer Science) will give an in-person colloquium on September 23rd 4th in the Seminar Room at T FS.
Title: Incorporating Commonsense Reasoning into
NLP Models
Abstract: NLP models are primarily superv
ised\, and are by design trained on a sample of the situations they may enc
ounter in practice. The ability of models to generalize to and address unkn
own situations reasonably is limited\, but may be improved by endowing mode
ls with commonsense knowledge and reasoning skills. In this talk\, I will p
resent several lines of work in which commonsense is used for improving the
performance of NLP tasks: for completing missing knowledge in underspecifi
ed language\, interpreting figurative language\, and resolving context-sens
itive event coreference. Finally\, I will discuss open problems and future
directions in building NLP models with commonsense reasoning abilities.
Bio: Vered Shwartz is an Assistant Professor of Compute r Science at the University of British Columbia and a faculty member at the Vector Institute for Artificial Intelligence. Her research interests inclu de commonsense reasoning\, computational semantics and pragmatics\, and mul tiword expressions. Previously\, Vered was a postdoctoral researcher at the Allen Institute for AI (AI2) and the University of Washington\, and receiv ed her PhD in Computer Science from Bar-Ilan University.
LOCATION:Totem Field Studios GEO:49.260872;-123.113952 URL;VALUE=URI:https://linguistics.ubc.ca/events/event/shwartz/ END:VEVENT BEGIN:VTIMEZONE TZID:America/Vancouver BEGIN:DAYLIGHT TZOFFSETFROM:-0800 TZOFFSETTO:-0700 DTSTART:20220313T100000 TZNAME:PDT END:DAYLIGHT END:VTIMEZONE END:VCALENDAR