Learning Kernel-Smoothed Machine Translation with Retrieved Examples

TL;DR

This paper introduces KSTER, a kernel-smoothed approach for online adaptation of neural machine translation models that improves translation quality without retraining, by effectively leveraging retrieved examples.

Contribution

The paper proposes a novel kernel-smoothed method for online NMT adaptation that outperforms existing methods without retraining.

Findings

Achieves 1.1 to 1.5 BLEU score improvements over existing methods.

Effective in domain adaptation and multi-domain translation tasks.

Does not require expensive retraining of models.

Abstract

How to effectively adapt neural machine translation (NMT) models according to emerging cases without retraining? Despite the great success of neural machine translation, updating the deployed models online remains a challenge. Existing non-parametric approaches that retrieve similar examples from a database to guide the translation process are promising but are prone to overfit the retrieved examples. In this work, we propose to learn Kernel-Smoothed Translation with Example Retrieval (KSTER), an effective approach to adapt neural machine translation models online. Experiments on domain adaptation and multi-domain machine translation datasets show that even without expensive retraining, KSTER is able to achieve improvement of 1.1 to 1.5 BLEU scores over the best existing online adaptation methods. The code and trained models are released at https://github.com/jiangqn/KSTER.