Context-Adaptive Document-Level Neural Machine Translation

TL;DR

This paper proposes a data-adaptive approach for document-level neural machine translation that dynamically selects relevant context, leading to significant performance improvements over fixed-context methods.

Contribution

Introduces a light predictor mechanism allowing models to adaptively select context, enhancing translation quality in document-level NMT.

Findings

Achieves up to 1.99 BLEU point improvement

Demonstrates effectiveness of adaptive context selection

Outperforms previous fixed-context models

Abstract

Most existing document-level neural machine translation (NMT) models leverage a fixed number of the previous or all global source sentences to handle the context-independent problem in standard NMT. However, the translating of each source sentence benefits from various sizes of context, and inappropriate context may harm the translation performance. In this work, we introduce a data-adaptive method that enables the model to adopt the necessary and useful context. Specifically, we introduce a light predictor into two document-level translation models to select the explicit context. Experiments demonstrate the proposed approach can significantly improve the performance over the previous methods with a gain up to 1.99 BLEU points.