Effective Strategies in Zero-Shot Neural Machine Translation
Ha, Thanh-Le 1; Niehues, Jan
1; Waibel, Alex 1
1 Institut für Anthropomatik und Robotik (IAR), Karlsruher Institut für Technologie (KIT)
1; Waibel, Alex 11 Institut für Anthropomatik und Robotik (IAR), Karlsruher Institut für Technologie (KIT)
Abstract:
In this paper, we proposed two strategies which can be applied to a multilingual neural machine translation system in order to better tackle zero-shot scenarios despite not having any parallel corpus. The experiments show that they are effective in terms of both performance and computing resources, especially in multilingual translation of unbalanced data in real zero-resourced condition when they alleviate the language bias problem.
| Zugehörige Institution(en) am KIT | Institut für Anthropomatik und Robotik (IAR) |
| Publikationstyp | Proceedingsbeitrag |
| Publikationsjahr | 2017 |
| Sprache | Englisch |
| Identifikator | KITopen-ID: 1000145055 |
| Erschienen in | Proceedings of the International Workshop on Spoken Language Translation 14th-15th December, 2017 Tokyo, Japan. Ed.: S. Sakti |
| Veranstaltung | 14th International Workshop on Spoken Language Translation (IWSLT 2017), Tokio, Japan, 14.12.2017 – 15.12.2017 |
| Verlag | ACL Anthology |
| Seiten | 105-112 |
| Externe Relationen | Abstract/Volltext Abstract/Volltext |
| Nachgewiesen in | arXiv |