switch-GLAT: Multilingual Parallel Machine Translation Via Code-Switch Decoder

Multilingual machine translation aims to develop a single model for multiple language directions. However, existing multilingual models based on Transformer are limited in terms of both translation performance and inference speed. In this paper, we propose switch-GLAT, a non-autoregressive multilingual machine translation model with a code-switch decoder. It can generate contextual code-switched translations for a given source sentence, and perform code-switch back-translation, greatly boosting multilingual translation performance. In addition, its inference is highly efficient thanks to its parallel decoder. Experiments show that our proposed switch-GLAT outperform the multilingual Transformer with as much as 0.74 BLEU improvement and 6.2x faster decoding speed in inference.

Authors

• Zhenqiao Song
• Hao Zhou 0012
• Lihua Qian
• Jingjing Xu 0001
• Shanbo Cheng
• Mingxuan Wang
• Lei Li 0005

Keywords

• multilingual non-autoregressive machine translation
• contextualized code-switching
• back-translation