Linear Cross-Lingual Mapping of Sentence Embeddings

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Vasilyev, Oleg, Isono, Fumika, Bohannon, John
Format: Preprint
Veröffentlicht: 2023
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866913405817847808
author Vasilyev, Oleg
Isono, Fumika
Bohannon, John
author_facet Vasilyev, Oleg
Isono, Fumika
Bohannon, John
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2305_14256
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Linear Cross-Lingual Mapping of Sentence Embeddings
Vasilyev, Oleg
Isono, Fumika
Bohannon, John
Computation and Language
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.
title Linear Cross-Lingual Mapping of Sentence Embeddings
topic Computation and Language
url https://arxiv.org/abs/2305.14256