Effective Strategies in Zero-Shot Neural Machine Translation

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.

arXiv commentsubmitted to IWSLT17

References in corpus (1)

• Adam: A Method for Stochastic Optimization

Cited by in corpus (16)

• The Missing Ingredient in Zero-Shot Neural Machine Translation
• OpenNMT: Neural Machine Translation Toolkit
• A Comprehensive Survey of Multilingual Neural Machine Translation
• A neural interlingua for multilingual machine translation
• Low-Resource Neural Machine Translation for Southern African Languages
• Improving Multilingual Neural Machine Translation For Low-Resource Languages: French,English - Vietnamese
• Subword Segmentation and a Single Bridge Language Affect Zero-Shot Neural Machine Translation
• Multilingual End-to-End Speech Translation
• Contrastive Learning for Many-to-many Multilingual Neural Machine Translation
• Translating Translationese: A Two-Step Approach to Unsupervised Machine Translation
• A Brief Survey of Multilingual Neural Machine Translation
• Neural Machine Translation: A Review and Survey
• Pivot-based Transfer Learning for Neural Machine Translation between Non-English Languages
• Machine Translation with Unsupervised Length-Constraints
• Controlling Neural Machine Translation Formality with Synthetic Supervision
• Zero-shot Speech Translation