Passer à la navigation principale Passer à la recherche Passer au contenu principal

DATScore: Evaluating Translation with Data Augmented Translations

  • École Polytechnique
  • Linagora
  • AUEB

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collectionContribution à une conférenceRevue par des pairs

Résumé

The rapid development of large pretrained language models has revolutionized not only the field of Natural Language Generation (NLG) but also its evaluation. Inspired by the recent work of BARTScore: a metric leveraging the BART language model to evaluate the quality of generated text from various aspects, we introduce DATScore. DATScore uses data augmentation techniques to improve the evaluation of machine translation. Our main finding is that introducing data augmented translations of the source and reference texts is greatly helpful in evaluating the quality of the generated translation. We also propose two novel score averaging and term weighting strategies to improve the original score computing process of BARTScore. Experimental results on WMT show that DATScore correlates better with human meta-evaluations than the other recent state-of-the-art metrics, especially for low-resource languages. Ablation studies demonstrate the value added by our new scoring strategies.

langue originaleAnglais
titreEACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2023
EditeurAssociation for Computational Linguistics (ACL)
Pages912-922
Nombre de pages11
ISBN (Electronique)9781959429470
Les DOIs
étatPublié - 1 janv. 2023
Modification externeOui
Evénement17th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2023 - Findings of EACL 2023 - Hybrid, Dubrovnik, Croatie
Durée: 2 mai 20236 mai 2023

Série de publications

NomEACL 2023 - 17th Conference of the European Chapter of the Association for Computational Linguistics, Findings of EACL 2023

Une conférence

Une conférence17th Conference of the European Chapter of the Association for Computational Linguistics, EACL 2023 - Findings of EACL 2023
Pays/TerritoireCroatie
La villeHybrid, Dubrovnik
période2/05/236/05/23

Empreinte digitale

Examiner les sujets de recherche de « DATScore: Evaluating Translation with Data Augmented Translations ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation