TY - GEN
T1 - Incorporating Phrase-Level Agreement into Neural Machine Translation
AU - Yang, Mingming
AU - Wang, Xing
AU - Zhang, Min
AU - Zhao, Tiejun
N1 - Publisher Copyright:
© 2020, Springer Nature Switzerland AG.
PY - 2020
Y1 - 2020
N2 - Phrase information has been successfully integrated into current state-of-the-art neural machine translation (NMT) models. However, the natural property of the source and target phrase alignment has not been explored. In this paper, we propose a novel phrase-level agreement method to deal with this problem. First, we explore n-gram models over minimal translation units (MTUs) to explicitly capture aligned bilingual phrases from the parallel corpora. Then, we propose a phrase-level agreement loss that directly reduces the difference between the representations of the source-side and target-side phrase. Finally, we integrate the phrase-level agreement loss into the NMT models, to improve the translation performance. Empirical results on the NIST Chinese-to-English and the WMT English-to-German translation tasks demonstrate that the proposed phrase-level agreement method achieves significant improvements over state-of-the-art baselines, demonstrating the effectiveness and necessity of exploiting phrase-level agreement for NMT.
AB - Phrase information has been successfully integrated into current state-of-the-art neural machine translation (NMT) models. However, the natural property of the source and target phrase alignment has not been explored. In this paper, we propose a novel phrase-level agreement method to deal with this problem. First, we explore n-gram models over minimal translation units (MTUs) to explicitly capture aligned bilingual phrases from the parallel corpora. Then, we propose a phrase-level agreement loss that directly reduces the difference between the representations of the source-side and target-side phrase. Finally, we integrate the phrase-level agreement loss into the NMT models, to improve the translation performance. Empirical results on the NIST Chinese-to-English and the WMT English-to-German translation tasks demonstrate that the proposed phrase-level agreement method achieves significant improvements over state-of-the-art baselines, demonstrating the effectiveness and necessity of exploiting phrase-level agreement for NMT.
KW - Minimal translation units
KW - Neural machine translation
KW - Phrase-level agreement
UR - https://www.scopus.com/pages/publications/85093095636
U2 - 10.1007/978-3-030-60450-9_33
DO - 10.1007/978-3-030-60450-9_33
M3 - 会议稿件
AN - SCOPUS:85093095636
SN - 9783030604493
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 416
EP - 428
BT - Natural Language Processing and Chinese Computing - 9th CCF International Conference, NLPCC 2020, Proceedings
A2 - Zhu, Xiaodan
A2 - Zhang, Min
A2 - Hong, Yu
A2 - He, Ruifang
PB - Springer Science and Business Media Deutschland GmbH
T2 - 9th CCF International Conference on Natural Language Processing and Chinese Computing, NLPCC 2020
Y2 - 14 October 2020 through 18 October 2020
ER -