Alibaba’s Neural Machine Translation Systems for WMT18

By Yongchao Deng, Shanbo Cheng, Jun Lu, Jingang Wang, Shenglan Wu, Liang Yao, Guchun Zhang, Haibo Zhang, Pei Zhang, Changfeng Zhu and 1 more. 2018

33 citations recorded by the source corpus.

This paper describes the submission systems of Alibaba for WMT18 shared news translation task. We participated in 5 translation directions including English ↔ Russian, English ↔ Turkish in both directions and English → Chinese. Our systems are based on Google’s Transformer model architecture, into which we integrated the most recent features from the academic research. We also employed most techniques that have been proven effective during the past WMT years, such as BPE, back translation, data selection, model ensembling and reranking, at industrial scale. For some morphologically-rich languages, we also incorporated linguistic knowledge into our neural network. For the translation tasks in which we have participated, our resulting systems achieved the best case sensitive BLEU score in all 5 directions. Notably, our English → Russian system outperformed the second reranked system by 5 BLEU score.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

Microsoft Research Asia’s Systems for WMT19 (2019, methodology): “We use the ﬁlter method in Deng et al. (2018) and continue to train the model on the ﬁltered data.”
The NiuTrans Machine Translation Systems for WMT19 (2019, methodology): “It has been proved effective in the past few years’ WMT tasks (Wang et al., 2018; Deng et al., 2018; Junczys-Dowmunt, 2018; Sennrich et al., 2016a).”
Findings of the 2018 Conference on Machine Translation (WMT18) (2018, methodology): “2.4.3 ALIBABA (Deng et al., 2018) Alibaba systems are based on the Transformer model architecture, with the most recent features from the academic research integrated, such as weighted Transformer, Transformer with relative position attention, etc.”
Simple and Effective Noisy Channel Modeling for Neural Machine Translation (2019, methodology): “We compare to various baselines including right-to-left sequence to sequence models which are a popular choice for re-ranking and regularly feature in successful WMT sub-5 missions (Deng et al., 2018; Koehn et al., 2018; Junczys-Dowmunt, 2018).”