Combating the Curse of Multilinguality in Cross-Lingual WSD by Aligning Sparse Contextualized Word RepresentationsDownload PDF


08 Mar 2022, 17:01 (modified: 02 May 2022, 21:46)NAACL 2022 Conference Blind SubmissionReaders: Everyone
Abstract: In this paper, we advocate for using large pre-trained monolingual language models in cross lingual zero-shot word sense disambiguation (WSD) coupled with a contextualized mapping mechanism. We also report rigorous experiments that illustrate the effectiveness of employing sparse contextualized word representations obtained via a dictionary learning procedure. Our experimental results demonstrate that the above modifications yield a significant improvement of nearly 6.5 points of increase in the average F-score (from 62.0 to 68.5) over a collection of 17 typologically diverse set of target languages. We release our source code for replicating our experiments at
Paper Link:
Paper Type: Long paper (up to eight pages of content + unlimited references and appendices)
Presentation Mode: This paper will be presented in person in Seattle
Copyright Consent Signature (type Name Or NA If Not Transferrable): Gábor Berend
Copyright Consent Name And Address: Institutie of Informatics, University of Szeged, 2. Árpád tér, Szeged, Hungary
0 Replies