BERTTune: Fine-Tuning Neural Machine Translation with BERTScore

Inigo Jauregi Unanue, Jacob Parnell, Massimo Piccardi


Abstract
Neural machine translation models are often biased toward the limited translation references seen during training. To amend this form of overfitting, in this paper we propose fine-tuning the models with a novel training objective based on the recently-proposed BERTScore evaluation metric. BERTScore is a scoring function based on contextual embeddings that overcomes the typical limitations of n-gram-based metrics (e.g. synonyms, paraphrases), allowing translations that are different from the references, yet close in the contextual embedding space, to be treated as substantially correct. To be able to use BERTScore as a training objective, we propose three approaches for generating soft predictions, allowing the network to remain completely differentiable end-to-end. Experiments carried out over four, diverse language pairs show improvements of up to 0.58 pp (3.28%) in BLEU score and up to 0.76 pp (0.98%) in BERTScore (F_BERT) when fine-tuning a strong baseline.
Anthology ID:
2021.acl-short.115
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers)
Month:
August
Year:
2021
Address:
Online
Editors:
Chengqing Zong, Fei Xia, Wenjie Li, Roberto Navigli
Venues:
ACL | IJCNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
915–924
Language:
URL:
https://aclanthology.org/2021.acl-short.115
DOI:
10.18653/v1/2021.acl-short.115
Bibkey:
Cite (ACL):
Inigo Jauregi Unanue, Jacob Parnell, and Massimo Piccardi. 2021. BERTTune: Fine-Tuning Neural Machine Translation with BERTScore. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 2: Short Papers), pages 915–924, Online. Association for Computational Linguistics.
Cite (Informal):
BERTTune: Fine-Tuning Neural Machine Translation with BERTScore (Jauregi Unanue et al., ACL-IJCNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-short.115.pdf
Video:
 https://aclanthology.org/2021.acl-short.115.mp4