Screening Gender Transfer in Neural Machine Translation

Fiche du document

Date

11 novembre 2021

Type de document
Périmètre
Langue
Identifiants
Collection

Archives ouvertes

Licences

http://creativecommons.org/licenses/by/ , info:eu-repo/semantics/OpenAccess




Citer ce document

Guillaume Wisniewski et al., « Screening Gender Transfer in Neural Machine Translation », HAL-SHS : linguistique, ID : 10670/1.5yjpo4


Métriques


Partage / Export

Résumé En

This paper aims at identifying the information flow in state-of-the-art machine translation systems, taking as example the transfer of gender when translating from French into English.Using a controlled set of examples, we experiment several ways to investigate how gender information circulates in a encoder-decoder architecture considering both probing techniques as well as interventions on the internal representations used in the MT system. Our results show that gender information can be found in all token representations built by the encoder and the decoder and lead us to conclude that there are multiple pathways for gender transfer.

document thumbnail

Par les mêmes auteurs

Sur les mêmes sujets

Sur les mêmes disciplines

Exporter en