Iterative Dual Domain Adaptation for Neural Machine Translation

TL;DR

This paper introduces an iterative dual domain adaptation framework for neural machine translation that repeatedly transfers knowledge between in-domain and out-of-domain models, improving translation quality across domains.

Contribution

It proposes a novel iterative bidirectional knowledge transfer method for domain adaptation in NMT, extending to multiple out-of-domain corpora based on domain similarity.

Findings

Improves translation performance on Chinese-English and English-German tasks.

Outperforms one-pass domain adaptation methods.

Effective in scenarios with multiple out-of-domain corpora.

Abstract

Previous studies on the domain adaptation for neural machine translation (NMT) mainly focus on the one-pass transferring out-of-domain translation knowledge to in-domain NMT model. In this paper, we argue that such a strategy fails to fully extract the domain-shared translation knowledge, and repeatedly utilizing corpora of different domains can lead to better distillation of domain-shared translation knowledge. To this end, we propose an iterative dual domain adaptation framework for NMT. Specifically, we first pre-train in-domain and out-of-domain NMT models using their own training corpora respectively, and then iteratively perform bidirectional translation knowledge transfer (from in-domain to out-of-domain and then vice versa) based on knowledge distillation until the in-domain NMT model convergences. Furthermore, we extend the proposed framework to the scenario of multiple out-of-domain training corpora, where the above-mentioned transfer is performed sequentially between the in-domain and each out-of-domain NMT models in the ascending order of their domain similarities. Empirical results on Chinese-English and English-German translation tasks demonstrate the effectiveness of our framework.