Conference paper Open Access

How Transformer Revitalizes Character-based Neural Machine Translation: An Investigation on Japanese-Vietnamese Translation Systems

Ngo, Thi-Vinh; Ha, Thanh-Le; Nguyen, Phuong-Thai; Nguyen, Le-Minh

While translating between East Asian languages, many works have discovered clear advantages of using characters as the translation unit. Unfortunately, traditional recurrent neural machine translation systems hinder the practical usage of those character-based systems due to their architectural limitations. They are unfavorable in handling extremely long sequences as well as highly restricted in parallelizing the computations. In this paper, we demonstrate that the new transformer architecture can perform character-based trans- lation better than the recurrent one. We conduct experiments on a low-resource language pair: Japanese-Vietnamese. Our models considerably outperform the state-of-the-art systems which employ word-based recurrent architectures.

Files (266.5 kB)
Name Size
IWSLT2019_paper_28.pdf
md5:5eecea8a3ca3a1487830b867ce8f9a4d
266.5 kB Download
55
38
views
downloads
All versions This version
Views 5556
Downloads 3839
Data volume 10.1 MB10.4 MB
Unique views 5354
Unique downloads 3738

Share

Cite as