返回
Preordering Encoding on Transformer for Translation
DOI:10.1109/TASLP.2020.3042001.png)
摘要
En 中文
The difference in word orders between source and target languages is a serious hurdle for machine translation. Preordering methods, which reorder the words in a source sentence before translation to obtain a similar word ordering with a target language, significantly improve the quality in statistical machine translation. While the information on the preordering position improved the translation quality in recurrent neural network-based models, questions such as how to use preordering information and whether it is helpful for the Transformer model remain unaddressed. In this article, we successfully employed preordering techniques in the Transformer-based neural machine translation. Specifically, we proposed a novel preordering encoding that exploits the reordering information of the source and target sentences as positional encoding in the Transformer model. Experimental results on ASPEC Japanese-English and WMT 2015 English-German, English-Czech, and English-Russian translation tasks confirmed that the proposed method significantly improved the translation quality evaluated by the BLEU scores of the Transformer model 1.34 points in the Japanese-to-English task, 2.19 points in the English-to-German task, 0.15 points in the Czech-to-English task, and 1.48 points in the English-to-Russian task.
Keyword:
Neural machine translation
preordering
word-order
transformer
期刊
I
IF:
5.1
论文数:
2.6K
被引数:
1.1W
机构
引用论文
Evaluations of antidepressant activity of Anacyclus pyrethrum root extractAnacyclus pyrethrum根提取物抗抑郁活性的评价
Teleseismic tomography across the middle Urals: lithospheric trace of an ancient continental collision中乌拉尔地区的远震层析成像:古老大陆碰撞的岩石圈痕迹
Acoustic stratigraphy and current-generated bedforms in deep ocean basins off southeastern Africa深海声学地层学和东南非洲大陆架外的流生底形
Effectiveness of winter temperatures for satisfying chilling requirements for reproductive budburst of red alder (Alnus rubra)冬季温度在满足红桤木(*Alnus rubra*)生殖芽萌发的需冷量方面的有效性
PeerJ
IF0

