When and Why are Pre-trained Word Embeddings Useful for Neural Machine Translation?

Abstract

The performance of Neural Machine Translation (NMT) systems often suffers in low-resource scenarios where sufficiently large-scale parallel corpora cannot be obtained. Pre-trained word embeddings have proven to be invaluable for improving performance in natural language analysis tasks, which often suffer from paucity of data. However, their utility for NMT has not been extensively explored. In this work, we perform five sets of experiments that analyze when we can expect pre-trained word embeddings to help in NMT tasks. We show that such embeddings can be surprisingly effective in some cases -- providing gains of up to 20 BLEU points in the most favorable setting.

arXiv commentNAACL 2018

References in corpus (2)

• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• XNMT: The eXtensible Neural Machine Translation Toolkit

Cited by in corpus (8)

• Deep Transformers with Latent Depth
• Transfer Learning in Multilingual Neural Machine Translation with Dynamic Vocabulary
• Incorporating Bilingual Dictionaries for Low Resource Semi-Supervised Neural Machine Translation
• Dynamic Knowledge embedding and tracing
• Robust Optimization for Multilingual Translation with Imbalanced Data
• Complexity of Symbolic Representation in Working Memory of Transformer Correlates with the Complexity of a Task
• Active Learning for Massively Parallel Translation of Constrained Text into Low Resource Languages
• A Dynamic Head Importance Computation Mechanism for Neural Machine Translation