A Survey of Deep Learning Techniques for Neural Machine Translation

TL;DR

This survey reviews the evolution, key developments, and research directions of Neural Machine Translation, highlighting its rapid growth and the diverse approaches within the field.

Contribution

It provides a comprehensive overview of NMT's development, categorizes research orientations, and discusses future trends, filling a gap in understanding the technology's progression.

Findings

NMT has rapidly evolved as a dominant machine translation approach.

Research in NMT covers various architectures and training methods.

Future trends include improving translation quality and model efficiency.

Abstract

In recent years, natural language processing (NLP) has got great development with deep learning techniques. In the sub-field of machine translation, a new approach named Neural Machine Translation (NMT) has emerged and got massive attention from both academia and industry. However, with a significant number of researches proposed in the past several years, there is little work in investigating the development process of this new technology trend. This literature survey traces back the origin and principal development timeline of NMT, investigates the important branches, categorizes different research orientations, and discusses some future research trends in this field.