Work

Disentangling Meaning and Language Components in Diverse Multilingual Sentence Embeddings

Year: 2026

Type: conference-paper

Abstract: We disentangle multilingual sentence embeddings into language-dependent and languageagnostic components, leveraging the latter to improve cross-lingual similarity estimation.Previous studies focused on encoder-based approaches that use only the input sentence; in contrast, this study examines the effectiveness of disentanglement methods across a broader range of sentence embeddings, including decoder-based approaches and those that utilize prompts.Experimental results demonstrate that embedding disentanglement is effective for a wide variety of sentence embeddings.

Source: Proceedings of the conference - Association for Computational Linguistics. Meeting

DOI: 10.18653/v1/2026.acl-srw.102

Authors: Kanade Nonomura, Keita Fukushima, Risa Kondo, Tomoyuki Kajiwara

Institutions: Ehime University, The University of Osaka

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: hybrid

OpenAlex ID: W7166886906

FWCI:

Cites:

Cited by:

Related to: 10