Focused Concatenation for Context-Aware Neural Machine Translation - Archive ouverte HAL Accéder directement au contenu
Communication Dans Un Congrès Année : 2022

Focused Concatenation for Context-Aware Neural Machine Translation

Lorenzo Lupo
Marco Dinarelli
Laurent Besacier
  • Fonction : Auteur
  • PersonId : 1106327

Résumé

A straightforward approach to context-aware neural machine translation consists in feeding the standard encoder-decoder architecture with a window of consecutive sentences, formed by the current sentence and a number of sentences from its context concatenated to it. In this work, we propose an improved concatenation approach that encourages the model to focus on the translation of the current sentence, discounting the loss generated by target context. We also propose an additional improvement that strengthen the notion of sentence boundaries and of relative sentence distance, facilitating model compliance to the context-discounted objective. We evaluate our approach with both average-translation quality metrics and contrastive test sets for the translation of inter-sentential discourse phenomena, proving its superiority to the vanilla concatenation approach and other sophisticated context-aware systems.
Fichier principal
Vignette du fichier
2022.wmt-1.77.pdf (606.96 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)

Dates et versions

hal-03930344 , version 1 (09-01-2023)

Identifiants

Citer

Lorenzo Lupo, Marco Dinarelli, Laurent Besacier. Focused Concatenation for Context-Aware Neural Machine Translation. Conference on Machine Translation, Association for Computational Linguistics, Dec 2022, Abu Dhabi, United Arab Emirates. pp.830-842. ⟨hal-03930344⟩

Collections

UGA MIAI ANR
16 Consultations
12 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More