Representational Isomorphism and Alignment of Multilingual Large Language Models

Open Access
Authors
Publication date 2024
Host editors
  • J. Sälevä
  • A. Owodunni
Book title The 4th Workshop on Multilingual Representation Learning : proceedings of the workshop
Book subtitle MRL 2024 : November 16, 2024
ISBN (electronic)
  • 9798891761841
Event 4th Workshop on Multilingual Representation Learning
Pages (from-to) 293-297
Number of pages 5
Publisher Kerrville, TX: Association for Computational Linguistics
Organisations
  • Faculty of Science (FNWI) - Informatics Institute (IVI)
Abstract In this extended abstract, we investigate the capability of Large Language Models (LLMs) to represent texts in multilingual contexts. Our findings reveal that sentence representations derived from LLMs exhibit a high degree of isomorphism across languages. This existing isomorphism facilitates representational alignments in few-shot settings. Specifically, by applying a contrastive objective at the representation level with only a small number (e.g., 100) of translation pairs, we significantly improve models’ performance on Semantic Textual Similarity (STS) tasks across languages.
Document type Conference contribution
Language English
Related publication Representational Isomorphism and Alignment of Multilingual Large Language Models
Published at https://doi.org/10.18653/v1/2024.mrl-1.24
Downloads
2024.mrl-1.24 (Final published version)
Permalink to this page
Back