Abstract
Multilingual language models were shown to allow for nontrivial transfer across scripts and languages. In this work, we study the structure of the internal representations that enable this transfer. We focus on the representation of gender distinctions as a practical case study, and examine the extent to which the gender concept is encoded in shared subspaces across different languages. Our analysis shows that gender representations consist of several prominent components that are shared across languages, alongside language-specific components. The existence of language-independent and language-specific components provides an explanation for an intriguing empirical observation we make: while gender classification transfers well across languages, interventions for gender removal, trained on a single language, do not transfer easily to others.
| Original language | English |
|---|---|
| Title of host publication | ACL 2022 - 7th Workshop on Representation Learning for NLP, RepL4NLP 2022 - Proceedings of the Workshop |
| Publisher | Association for Computational Linguistics (ACL) |
| Pages | 67-77 |
| Number of pages | 11 |
| ISBN (Electronic) | 9781955917483 |
| State | Published - 2022 |
| Event | 7th Workshop on Representation Learning for NLP, RepL4NLP 2022 at ACL 2022 - Dublin, Ireland Duration: 26 May 2022 → … |
Publication series
| Name | Proceedings of the Annual Meeting of the Association for Computational Linguistics |
|---|---|
| ISSN (Print) | 0736-587X |
Conference
| Conference | 7th Workshop on Representation Learning for NLP, RepL4NLP 2022 at ACL 2022 |
|---|---|
| Country/Territory | Ireland |
| City | Dublin |
| Period | 26/05/22 → … |
Bibliographical note
Publisher Copyright:© 2022 Association for Computational Linguistics.