COOKIES: By using this website you agree that we can place Google Analytics Cookies on your device for performance monitoring. |
University of Cambridge > Talks.cam > Language Technology Lab Seminars > Probing Language Models for Paraphrastic Representations of Negation and Antonymy
Probing Language Models for Paraphrastic Representations of Negation and AntonymyAdd to your list(s) Download to your calendar using vCal
If you have a question about this talk, please contact Panagiotis Fytas. The ability of large language models to represent essential linguistic properties makes them applicable across a wide range of tasks. However, it is still an open question how well they address systematic compositionality and to what level of abstraction they reflect the actual meaning behind a given sentence. I present work on constructing a dedicated challenge set focusing on a specific linguistic phenomenon. In particular, the challenge set assesses the ability of language models to properly represent paraphrasticity realised in expressions that incorporate antonymy and negation. I present a large-scale evaluation of publicly available pre-trained language models on the proposed challenge set, showing large variation between models fine-tuned for different downstream tasks. This talk is part of the Language Technology Lab Seminars series. This talk is included in these lists:
Note that ex-directory lists are not shown. |
Other listsHome Murray Edwards College Politics Society sureapksOther talksModeling crossmodal attention in humanoid robots for HRI in complex social scenarios Data-based reductions and closure models for dynamical systems Programmable Emulsions By Design A scientists guide to the art of radiation therapy & Neurosurgery for the oncologist New fields in the strong gravity regime Difficulties of Seeing Through Walls: Why Stand-off Structure Reconnaissance with Radar is Hard |