Context-aware Decoder for Neural Machine Translation using a Target-side Document-Level Language Model

TL;DR

This paper introduces a simple method to enhance neural machine translation by integrating a target-side document-level language model into the decoder, enabling context-aware translation without requiring document-level parallel data.

Contribution

The paper proposes a novel approach to incorporate document-level context into sentence-level translation models using a target-side language model and mutual information representation.

Findings

Improved BLEU scores in three language pairs

Effective context-aware translation demonstrated in contrastive tests

No need for document-level parallel corpora

Abstract

Although many context-aware neural machine translation models have been proposed to incorporate contexts in translation, most of those models are trained end-to-end on parallel documents aligned in sentence-level. Because only a few domains (and language pairs) have such document-level parallel data, we cannot perform accurate context-aware translation in most domains. We therefore present a simple method to turn a sentence-level translation model into a context-aware model by incorporating a document-level language model into the decoder. Our context-aware decoder is built upon only a sentence-level parallel corpora and monolingual corpora; thus no document-level parallel data is needed. In a theoretical viewpoint, the core part of this work is the novel representation of contextual information using point-wise mutual information between context and the current sentence. We show the effectiveness of our approach in three language pairs, English to French, English to Russian, and Japanese to English, by evaluation in \textsc{bleu} and contrastive tests for context-aware translation.