Modeling Dual Read/Write Paths for Simultaneous Machine Translation

Shaolei Zhang, Yang Feng


Abstract
Simultaneous machine translation (SiMT) outputs translation while reading source sentence and hence requires a policy to decide whether to wait for the next source word (READ) or generate a target word (WRITE), the actions of which form a read/write path. Although the read/write path is essential to SiMT performance, no direct supervision is given to the path in the existing methods. In this paper, we propose a method of dual-path SiMT which introduces duality constraints to direct the read/write path. According to duality constraints, the read/write path in source-to-target and target-to-source SiMT models can be mapped to each other. As a result, the two SiMT models can be optimized jointly by forcing their read/write paths to satisfy the mapping. Experiments on En-Vi and De-En tasks show that our method can outperform strong baselines under all latency.
Anthology ID:
2022.acl-long.176
Volume:
Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
Month:
May
Year:
2022
Address:
Dublin, Ireland
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2461–2477
Language:
URL:
https://aclanthology.org/2022.acl-long.176
DOI:
10.18653/v1/2022.acl-long.176
Bibkey:
Cite (ACL):
Shaolei Zhang and Yang Feng. 2022. Modeling Dual Read/Write Paths for Simultaneous Machine Translation. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 2461–2477, Dublin, Ireland. Association for Computational Linguistics.
Cite (Informal):
Modeling Dual Read/Write Paths for Simultaneous Machine Translation (Zhang & Feng, ACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.acl-long.176.pdf
Code
 ictnlp/dual-paths