Exploiting Sentential Context for Neural Machine Translation

TL;DR

This paper introduces novel methods to incorporate sentential context at different encoder layers in neural machine translation, significantly improving translation quality over existing models like Transformer.

Contribution

It proposes shallow and deep sentential context extraction techniques that enhance encoder representations for better translation performance.

Findings

Improved translation results on WMT14 benchmarks

Deep sentential context outperforms shallow context

Consistent performance gains over Transformer model

Abstract

In this work, we present novel approaches to exploit sentential context for neural machine translation (NMT). Specifically, we first show that a shallow sentential context extracted from the top encoder layer only, can improve translation performance via contextualizing the encoding representations of individual words. Next, we introduce a deep sentential context, which aggregates the sentential context representations from all the internal layers of the encoder to form a more comprehensive context representation. Experimental results on the WMT14 English-to-German and English-to-French benchmarks show that our model consistently improves performance over the strong TRANSFORMER model (Vaswani et al., 2017), demonstrating the necessity and effectiveness of exploiting sentential context for NMT.