Measuring Lexico-Semantic Alignment in Debates with Contextualized Word Representations

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Dialog participants sometimes align their linguistic styles, e.g., they use the same words and syntactic constructions as their interlocutors. We propose to investigate the notion of lexico-semantic alignment: to what extent do speakers convey the same meaning when they use the same words? We design measures of lexico-semantic alignment relying on contextualized word representations. We show that they reflect interesting semantic differences between the two sides of a debate and that they can assist in the task of debate’s winner prediction.

Original languageEnglish
Title of host publication1st Workshop on Social Influence in Conversations, SICon 2023 - Proceedings of the Workshop
EditorsKushal Chawla, Weiyan Shi
PublisherAssociation for Computational Linguistics (ACL)
Pages50-63
Number of pages14
ISBN (Electronic)9781959429784
Publication statusPublished - 1 Jan 2023
Event1st Workshop on Social Influence in Conversations, SICon 2023, co-located with ACL 2023 - Toronto, Canada
Duration: 14 Jul 2023 → …

Publication series

NameProceedings of the Annual Meeting of the Association for Computational Linguistics
ISSN (Print)0736-587X

Conference

Conference1st Workshop on Social Influence in Conversations, SICon 2023, co-located with ACL 2023
Country/TerritoryCanada
CityToronto
Period14/07/23 → …

Fingerprint

Dive into the research topics of 'Measuring Lexico-Semantic Alignment in Debates with Contextualized Word Representations'. Together they form a unique fingerprint.

Cite this