BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:CIS - "Get to know your neighbors" Seminar series - Prof. Antoine 
 Bosselut
DTSTART:20220328T151500
DTEND:20220328T161500
DTSTAMP:20260916T045154Z
UID:2137d6dbdcc5f1cd24851ffce3f381add7c7337b7a2d400a0bb3b25e
CATEGORIES:Conferences - Seminars
DESCRIPTION:Prof. Antoine Bosselut\nTitle: Neuro-symbolic Scaffolds for Co
 mmonsense Representation and Reasoning\n\nAbstract: Natural language unde
 rspecifies the information-rich situations that it describes and communi
 cates. For example\, the sentence "She pumped her fist" connotes many pote
 ntial auspicious causes. For machines to understand natural language\, the
 y must be able to make commonsense inferences that enrich explicitly state
 d information. However\, current NLP systems lack the ability to ground th
 e situations they read and write about to relevant world knowledge. Moreov
 er\, they struggle to reason over available facts to generalize to future 
 unseen events. In this talk\, I will describe efforts at transforming mode
 rn language models into commonsense knowledge models by leveraging implici
 tly encoded knowledge representations. Then\, I will discuss work in desig
 ning natural language reasoning systems that use knowledge graphs as a str
 uctural scaffold for aggregating information across relevant commonsense i
 nferences.\n\nBio: Antoine Bosselut is an assistant professor in the Scho
 ol of Computer and Communication Sciences at the École Polytechnique Féd
 éral de Lausanne (EPFL) and the head of the Natural Language Processing (
 NLP) lab. Previously\, he was a postdoctoral fellow at Stanford University
  and a Young Investigator at the Allen Institute for AI (AI2). He received
  his PhD at the University of Washington. He was named as a Forbes 30 unde
 r 30 for Science and Healthcare in Europe (2021). His research focuses on 
 building knowledge-aware NLP systems\, specializing in commonsense represe
 ntation and reasoning.\n\n\nThe Center for Intelligent Systems at EPFL (CI
 S) is a collaboration among IC\, ENAC\, SB\; SV and STI that brings togeth
 er researchers working on different aspects of Intelligent Systems.\nIn or
 der to promote exchanges among researchers and encourage the creation of n
 ew\, collaborative projects\, CIS is organizing a "Get to know your neighb
 ors" series. Each seminar will consist of one short overview presentation 
 geared to the general public at EPFL.   \n \nThe CIS seminar will take
  place In hybrid mode: Room INF 328 and by Zoom:  https://epfl.zoom.us/j/
 64002897669\n\nPlease connect to your zoom account using your "@epfl.ch" a
 ddress\, as this live event is only open to the EPFL community\nMonday\, M
 arch 28th\, 2022 from 3:15 to 4:15 pm\nNB: Video recordings of the seminar
 s will be made available on our website and published on our social media 
 pages
LOCATION:INF 328 https://plan.epfl.ch/?room==INF%20328 https://epfl.zoom.u
 s/j/64002897669
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR
