One Sentence One Model for Neural Machine Translation
This addresses the limitation of fixed NMT models for specific test sentences, offering a domain-specific incremental improvement for machine translation tasks.
The paper tackled the problem of neural machine translation (NMT) by proposing a dynamic approach that fine-tunes a general network for each test sentence using similar bilingual data, resulting in significant translation performance improvements, especially with highly similar sentences.
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.