Diverse Pretrained Context Encodings Improve Document Translation

Domenic Donato, Lei Yu, Chris Dyer


Abstract
We propose a new architecture for adapting a sentence-level sequence-to-sequence transformer by incorporating multiple pre-trained document context signals and assess the impact on translation performance of (1) different pretraining approaches for generating these signals, (2) the quantity of parallel data for which document context is available, and (3) conditioning on source, target, or source and target contexts. Experiments on the NIST Chinese-English, and IWSLT and WMT English-German tasks support four general conclusions: that using pre-trained context representations markedly improves sample efficiency, that adequate parallel data resources are crucial for learning to use document context, that jointly conditioning on multiple context representations outperforms any single representation, and that source context is more valuable for translation performance than target side context. Our best multi-context model consistently outperforms the best existing context-aware transformers.
Anthology ID:
2021.acl-long.104
Volume:
Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long 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:
1299–1311
Language:
URL:
https://aclanthology.org/2021.acl-long.104
DOI:
10.18653/v1/2021.acl-long.104
Bibkey:
Cite (ACL):
Domenic Donato, Lei Yu, and Chris Dyer. 2021. Diverse Pretrained Context Encodings Improve Document Translation. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 1299–1311, Online. Association for Computational Linguistics.
Cite (Informal):
Diverse Pretrained Context Encodings Improve Document Translation (Donato et al., ACL-IJCNLP 2021)
Copy Citation:
PDF:
https://aclanthology.org/2021.acl-long.104.pdf
Video:
 https://aclanthology.org/2021.acl-long.104.mp4