BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Language Technology Lab Seminars
SUMMARY:Towards Knowledge-Robust and Multimodally-Grounded
  NLP - Mohit Bansal\, UNC Chapel Hill
DTSTART;TZID=Europe/London:20210610T170000
DTEND;TZID=Europe/London:20210610T180000
UID:TALK160972AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/160972
DESCRIPTION:In this talk\, I will present our group's work on 
 NLP models that are knowledge-robust and multimoda
 lly-grounded. First\, we will describe multi-task 
 and reinforcement learning methods to incorporate 
 novel auxiliary-skill tasks such as saliency\, ent
 ailment\, and back-translation validity (including
  bandit-based methods for automatic auxiliary task
  selection+mixing and multi-reward mixing). Next\,
  we will discuss developing adversarial robustness
  against reasoning shortcuts\, missing commonsense
  gaps\, and cross-domain/lingual generalization in
  QA and dialogue models (including auto-adversary 
 generation). Lastly\, we will discuss multimodally
 -grounded models which condition and reason on dyn
 amic spatio-temporal information in images and vid
 eos\, and action-based robotic navigation and asse
 mbling tasks (including commonsense reasoning for 
 ambiguous robotic instructions).
LOCATION:https://cam-ac-uk.zoom.us/j/97599459216?pwd=QTRsOW
 ZCOXRTREVnbTJBdXVpOXFvdz09
CONTACT:Marinela Parovic
END:VEVENT
END:VCALENDAR