Neural System Combination for Machine Translation
This work addresses the translation quality gap for users of machine translation systems by proposing an incremental improvement through system combination.
The paper tackles the problem of combining neural machine translation (NMT) and statistical machine translation (SMT) to leverage their respective strengths in fluency and adequacy, resulting in a model that achieves a 5.3 BLEU point improvement over the best single system and 3.4 BLEU points over state-of-the-art traditional methods on Chinese-to-English translation.
Neural machine translation (NMT) becomes a new approach to machine translation and generates much more fluent results compared to statistical machine translation (SMT). However, SMT is usually better than NMT in translation adequacy. It is therefore a promising direction to combine the advantages of both NMT and SMT. In this paper, we propose a neural system combination framework leveraging multi-source NMT, which takes as input the outputs of NMT and SMT systems and produces the final translation. Extensive experiments on the Chinese-to-English translation task show that our model archives significant improvement by 5.3 BLEU points over the best single system output and 3.4 BLEU points over the state-of-the-art traditional system combination methods.