Dual Past and Future for Neural Machine Translation

TL;DR

This paper introduces a dual framework for neural machine translation that uses both source-to-target and target-to-source models to better model Past and Future context, improving translation adequacy.

Contribution

It proposes a novel dual approach that directly supervises Past and Future modules using bidirectional NMT models, enhancing translation quality.

Findings

Significant improvement in translation adequacy.

Outperforms previous methods on benchmark tasks.

Effective modeling of Past and Future contexts.

Abstract

Though remarkable successes have been achieved by Neural Machine Translation (NMT) in recent years, it still suffers from the inadequate-translation problem. Previous studies show that explicitly modeling the Past and Future contents of the source sentence is beneficial for translation performance. However, it is not clear whether the commonly used heuristic objective is good enough to guide the Past and Future. In this paper, we present a novel dual framework that leverages both source-to-target and target-to-source NMT models to provide a more direct and accurate supervision signal for the Past and Future modules. Experimental results demonstrate that our proposed method significantly improves the adequacy of NMT predictions and surpasses previous methods in two well-studied translation tasks.