Linear Cross-Lingual Mapping of Sentence Embeddings

Oleg Vasilyev, Fumika Isono, John Bohannon


Abstract
Semantics of a sentence is defined with much less ambiguity than semantics of a single word, and we assume that it should be better preserved by translation to another language. If multilingual sentence embeddings intend to represent sentence semantics, then the similarity between embeddings of any two sentences must be invariant with respect to translation. Based on this suggestion, we consider a simple linear cross-lingual mapping as a possible improvement of the multilingual embeddings. We also consider deviation from orthogonality conditions as a measure of deficiency of the embeddings.
Anthology ID:
2024.findings-acl.486
Volume:
Findings of the Association for Computational Linguistics ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand and virtual meeting
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
8163–8171
Language:
URL:
https://aclanthology.org/2024.findings-acl.486
DOI:
Bibkey:
Cite (ACL):
Oleg Vasilyev, Fumika Isono, and John Bohannon. 2024. Linear Cross-Lingual Mapping of Sentence Embeddings. In Findings of the Association for Computational Linguistics ACL 2024, pages 8163–8171, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.
Cite (Informal):
Linear Cross-Lingual Mapping of Sentence Embeddings (Vasilyev et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2024.findings-acl.486.pdf