Jointly Learning to Align and Translate with Transformer Models

Sarthak Garg, Stephan Peitz, Udhyakumar Nallasamy, Matthias Paulik


Abstract
The state of the art in machine translation (MT) is governed by neural approaches, which typically provide superior translation accuracy over statistical approaches. However, on the closely related task of word alignment, traditional statistical word alignment models often remain the go-to solution. In this paper, we present an approach to train a Transformer model to produce both accurate translations and alignments. We extract discrete alignments from the attention probabilities learnt during regular neural machine translation model training and leverage them in a multi-task framework to optimize towards translation and alignment objectives. We demonstrate that our approach produces competitive results compared to GIZA++ trained IBM alignment models without sacrificing translation accuracy and outperforms previous attempts on Transformer model based word alignment. Finally, by incorporating IBM model alignments into our multi-task training, we report significantly better alignment accuracies compared to GIZA++ on three publicly available data sets.
Anthology ID:
D19-1453
Volume:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Month:
November
Year:
2019
Address:
Hong Kong, China
Editors:
Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
Venues:
EMNLP | IJCNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
4453–4462
Language:
URL:
https://aclanthology.org/D19-1453
DOI:
10.18653/v1/D19-1453
Bibkey:
Cite (ACL):
Sarthak Garg, Stephan Peitz, Udhyakumar Nallasamy, and Matthias Paulik. 2019. Jointly Learning to Align and Translate with Transformer Models. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 4453–4462, Hong Kong, China. Association for Computational Linguistics.
Cite (Informal):
Jointly Learning to Align and Translate with Transformer Models (Garg et al., EMNLP-IJCNLP 2019)
Copy Citation:
PDF:
https://aclanthology.org/D19-1453.pdf
Attachment:
 D19-1453.Attachment.zip
Code
 pytorch/fairseq