Fine-tuning mBART50 with French and Farsi data to improve the translation of Farsi dislocations into English and French - Université Paris Cité Accéder directement au contenu
Communication Dans Un Congrès Année : 2023

Fine-tuning mBART50 with French and Farsi data to improve the translation of Farsi dislocations into English and French

Résumé

In this paper, we discuss the improvements brought by the fine-tuning of mBART50 for the translation of a specific Farsi dataset of dislocations. Given our BLEU scores, our evaluation is mostly qualitative: we assess the improvements of our fine-tuning in the translations into French of our test dataset of Farsi. We describe the fine-tuning procedure and discuss the quality of the results in the translations from Farsi. We assess the sentences in the French translations that contain English tokens and for the English translations, we examine the ability of the finetuned system to translate Farsi dislocations into English without replicating the dislocated item as a double subject. We scrutinized the Farsi training data used to train for mBART50 (Tang et al., 2021). We fine-tuned mBART50 with samples from an in-house French-Farsi aligned translation of a short story. In spite of the scarcity of available resources, we found that finetuning with aligned French-Farsi data dramatically improved the grammatical well-formedness of the predictions for French, even if serious semantic issues remained. We replicated the experiment with the English translation of the same Farsi short story for a Farsi-English fine-tuning and found out that similar semantic inadequacies cropped up, and that some translations were worse than our mBART50 baseline. We showcased the fine-tuning of mBART50 with supplementary data and discussed the asymmetry of the situation, adding little data in the fine-tuning is sufficient to improve morpho-syntax for one language pair but seems to degrade translation to English.
Fichier principal
Vignette du fichier
MT_Summit_2023_MBART50_Users__Track (1).pdf (98.81 Ko) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Licence : CC BY NC - Paternité - Pas d'utilisation commerciale

Dates et versions

hal-04329162 , version 1 (07-12-2023)

Identifiants

  • HAL Id : hal-04329162 , version 1

Citer

Behnoosh Namdarzadeh, Sadaf Mohseni, Lichao Zhu, Guillaume Wisniewski, Nicolas Ballier. Fine-tuning mBART50 with French and Farsi data to improve the translation of Farsi dislocations into English and French. Machine Translation Summit 2023, University of Macau; NLP2CT lab, University of Macau; AAMT; MACETS, Sep 2023, Macau SAR, China. pp.152-161. ⟨hal-04329162⟩
18 Consultations
6 Téléchargements

Partager

Gmail Facebook X LinkedIn More