Synchronously Generating Two Languages with Interactive Decoding

Yining Wang, Jiajun Zhang, Long Zhou, Yuchen Liu, Chengqing Zong


Abstract
In this paper, we introduce a novel interactive approach to translate a source language into two different languages simultaneously and interactively. Specifically, the generation of one language relies on not only previously generated outputs by itself, but also the outputs predicted in the other language. Experimental results on IWSLT and WMT datasets demonstrate that our method can obtain significant improvements over both conventional Neural Machine Translation (NMT) model and multilingual NMT model.
Anthology ID:
D19-1330
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:
3350–3355
Language:
URL:
https://aclanthology.org/D19-1330
DOI:
10.18653/v1/D19-1330
Bibkey:
Cite (ACL):
Yining Wang, Jiajun Zhang, Long Zhou, Yuchen Liu, and Chengqing Zong. 2019. Synchronously Generating Two Languages with Interactive Decoding. 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 3350–3355, Hong Kong, China. Association for Computational Linguistics.
Cite (Informal):
Synchronously Generating Two Languages with Interactive Decoding (Wang et al., EMNLP-IJCNLP 2019)
Copy Citation:
PDF:
https://aclanthology.org/D19-1330.pdf