LIT Team’s System Description for Japanese-Chinese Machine Translation Task in IWSLT 2020

Proceedings of the 17th International Conference on Spoken Language Translation (IWSLT), pages 109–113 July 9-10, 2020. c⃝2020 Association for Computational Linguistics https://doi.org/10.18653/v1/P17 109 LIT Team’s System Description for Japanese-Chinese Machine Translation Task in IWSLT 2020 Yimeng Zhuang, Yuan Zhang, Lijie Wang Samsung Research China - Beijing (SRC-B) {ym.zhuang, yuan.zhang, lijie.wang}@samsung.com Abstract This paper describes the LIT Team’s submis- sion to the IWSLT2020 open domain transla- tion task, focusing primarily on Japanese-to- Chinese translation direction. Our system is based on the organizers’ baseline system, but we do more works on improving the Trans- former baseline system by elaborate data pre- processing. We manage to obtain significant improvements, and this paper aims to share some data processing experiences in this trans- lation task. Large-scale back-translation on monolingual corpus is also investigated. In ad- dition, we also try shared and exclusive word embeddings, compare different granularity of tokens like sub-word level. Our Japanese-to- Chinese translation system achieves a perfor- mance of BLEU=34.0 and ranks 2nd among all participating systems. 1 Introduction In recent years, the neural machine translation (NMT) (Sun et al., 2019; Wu et al., 2016; Sen- nrich et al., 2015) has made great progress based on encoder-decoder architecture. We participate in the IWSLT 2020 open domain translation task: Japanese-to-Chinese. This paper describes the NMT systems for the IWSLT 2020 Japanese-to- Chinese machine translation task (Ansari et al., 2020). Our main efforts are data pre-processing, specifi- cally parallel data filter and sentence alignment. By elaborate data processing, we successfully improve the quality of the training set and thus boost the performance of our translation system. The back- translation mechanism (Edunov et al., 2018) is also investigated to extend the training corpus, we trans- late Chinese to Japanese to get the Japanese-to- Chinese training corpus, it is an effective approach to exploit the corresponding monolingual data sets. The transformer model (Vaswani et al., 2017) based on multi-head attention has achieved excel- lent performance in a variety of neural machine translation tasks in the last three years. This kind of NMT model surpasses the performance of the traditional statistical machine translation and the NMT performs particularly well especially with rich resource corpus. In our system, we adopt big- ger transformer architecture, since the performance of the Transformer relies on model capacity, ex. the number of dimensions of the feed-forward network. To improve performance, we adopted the Rela- tive Position Attention (Shaw et al., 2018). Also, we conduct experiments to compare the shared source and target word embeddings and exclu- sive word embeddings, and whether to adopt the shared embeddings relates to the translation direc- tion. The Chinese-to-Japanese direction achieves higher scores when adopting shared word embed- dings, however, the Japanese-to-Chinese direction produces opposite results. The paper is structured as follows: Section 2 will present a detailed description of our data pre- processing, back-translation is introduced in Sec- tion 3, the model of our system will be introduced in Section 4, the main results of our experiment will be shown in Section 5. Section 6 …

Content truncated to 500 words