End-to-End Neural Word Alignment Outperforms GIZA++

Thomas Zenkel, Joern Wuebker, John DeNero


Abstract
Word alignment was once a core unsupervised learning task in natural language processing because of its essential role in training statistical machine translation (MT) models. Although unnecessary for training neural MT models, word alignment still plays an important role in interactive applications of neural machine translation, such as annotation transfer and lexicon injection. While statistical MT methods have been replaced by neural approaches with superior performance, the twenty-year-old GIZA++ toolkit remains a key component of state-of-the-art word alignment systems. Prior work on neural word alignment has only been able to outperform GIZA++ by using its output during training. We present the first end-to-end neural word alignment method that consistently outperforms GIZA++ on three data sets. Our approach repurposes a Transformer model trained for supervised translation to also serve as an unsupervised word alignment model in a manner that is tightly integrated and does not affect translation quality.
Anthology ID:
2020.acl-main.146
Volume:
Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics
Month:
July
Year:
2020
Address:
Online
Editors:
Dan Jurafsky, Joyce Chai, Natalie Schluter, Joel Tetreault
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
1605–1617
Language:
URL:
https://aclanthology.org/2020.acl-main.146
DOI:
10.18653/v1/2020.acl-main.146
Bibkey:
Cite (ACL):
Thomas Zenkel, Joern Wuebker, and John DeNero. 2020. End-to-End Neural Word Alignment Outperforms GIZA++. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 1605–1617, Online. Association for Computational Linguistics.
Cite (Informal):
End-to-End Neural Word Alignment Outperforms GIZA++ (Zenkel et al., ACL 2020)
Copy Citation:
PDF:
https://aclanthology.org/2020.acl-main.146.pdf
Video:
 http://slideslive.com/38929138