BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Investigating Reasons for Disagreement in Natural Language Inferen
 ce - Marie-Catherine de Marneffe (FNRS – UCLouvain – The Ohio State Un
 iversity)
DTSTART:20230609T110000Z
DTEND:20230609T120000Z
UID:TALK200038@talks.cam.ac.uk
CONTACT:Michael Schlichtkrull
DESCRIPTION:Abstract:\n\nCurrent practices of operationalizing annotations
  in crowdsourced datasets for natural language understanding (NLU) too oft
 en assume one single label per item. In this talk\, I argue that NLU shoul
 d investigate disagreement in annotations – human label variation (Plank
  2022) – to fully capture human interpretations of language. I investiga
 te how human label variation in natural language inference (NLI) arises\, 
 focusing on linguistic phenomena present in the sentences that lead to dif
 ferent interpretations. I also explore two modeling approaches for detecti
 ng items with potential disagreement (a 4-way classification with a Compli
 cated label in addition to the three standard NLI labels\, and a multilabe
 l classification approach)\, and evaluate whether these approaches recall 
 the possible interpretations in the data.\n\nBio:\n\nMarie-Catherine de Ma
 rneffe obtained her PhD from Stanford University in 2012. She is an associ
 ate professor in Linguistics at The Ohio State University. She also got ap
 pointed as a FNRS Research Associate at UCLouvain in 2022. Her research fo
 cuses on developing computational linguistic methods that capture what is 
 conveyed by language beyond the literal meaning of the words. In particula
 r she works on "veridicality": how do people interpret events they read ab
 out in the news -- do they think such events really happen\, did not happe
 n\, or are just a possibility? Primarily she wants to ground meanings in c
 orpus data and show how such meanings can drive pragmatic inference. She h
 as also contributed to defining the Stanford Dependencies and the Universa
 l Dependencies representations. Her research has been funded by Google Inc
 . and the NSF.
LOCATION:Computer Lab\, FW26
END:VEVENT
END:VCALENDAR