Modelling Lexical Ambiguity with Density Matrices

Francois Meyer, Martha Lewis

Research output: Chapter in Book/Report/Conference proceedingConference Contribution (Conference Proceeding)

Abstract

Words can have multiple senses. Compositional distributional models of meaning have been argued to deal well with finer shades of meaning variation known as polysemy, but are not so well equipped to handle word senses that are etymologically unrelated, or homonymy. Moving from vectors to density matrices allows us to encode a probability distribution over different senses of a word, and can also be accommodated within a compositional distributional model of meaning. In this paper we present three new neural models for learning density matrices from a corpus, and test their ability to discriminate between word senses on a range of compositional datasets. When paired with a particular composition method, our best model outperforms existing vector-based compositional models as well as strong sentence encoders.
Original languageEnglish
Title of host publicationProceedings of the 24th Conference on Computational Natural Language Learning
PublisherAssociation for Computational Linguistics
Pages276-290
Number of pages15
DOIs
Publication statusPublished - 19 Nov 2020
Event24th Conference on Computational Natural Language Learning - Online
Duration: 19 Nov 202020 Nov 2020
https://www.conll.org/2020

Conference

Conference24th Conference on Computational Natural Language Learning
Abbreviated titleCoNLL 2020
Period19/11/2020/11/20
Internet address

Cite this