One Sentence One Model for Neural Machine Translation

TL;DR

This paper introduces a dynamic neural machine translation approach that fine-tunes the model for each test sentence using similar training data, significantly enhancing translation quality.

Contribution

It proposes a novel dynamic NMT method that adapts the model to each test sentence via fine-tuning on similar data, improving translation accuracy.

Findings

Significant translation improvements with similar sentence data

Effective fine-tuning enhances NMT performance

Method outperforms fixed models in experiments

Abstract

Neural machine translation (NMT) becomes a new state-of-the-art and achieves promising translation results using a simple encoder-decoder neural network. This neural network is trained once on the parallel corpus and the fixed network is used to translate all the test sentences. We argue that the general fixed network cannot best fit the specific test sentences. In this paper, we propose the dynamic NMT which learns a general network as usual, and then fine-tunes the network for each test sentence. The fine-tune work is done on a small set of the bilingual training data that is obtained through similarity search according to the test sentence. Extensive experiments demonstrate that this method can significantly improve the translation performance, especially when highly similar sentences are available.