Cross-lingual Word Embeddings beyond Zero-shot Machine Translation

Abstract

We explore the transferability of a multilingual neural machine translation model to unseen languages when the transfer is grounded solely on the cross-lingual word embeddings. Our experimental results show that the translation knowledge can transfer weakly to other languages and that the degree of transferability depends on the languages' relatedness. We also discuss the limiting aspects of the multilingual architectures that cause weak translation transfer and suggest how to mitigate the limitations.

Accepted at the 8th Swedish Language Technology Conference (SLTC-2020)

References in corpus (9)

• Adam: A Method for Stochastic Optimization
• Neural Machine Translation by Jointly Learning to Align and Translate
• Exploiting Similarities among Languages for Machine Translation
• Enriching Word Vectors with Subword Information
• Massively Multilingual Word Embeddings
• Word Translation Without Parallel Data
• Transfer Learning across Low-Resource, Related Languages for Neural Machine Translation
• XNMT: The eXtensible Neural Machine Translation Toolkit
• Improving Zero-Shot Translation of Low-Resource Languages