Encoding Source Language with Convolutional Neural Network for Machine Translation
Abstract
The recently proposed neural network joint model (NNJM) (Devlin et al., 2014) augments the n-gram target language model with a heuristically chosen source context window, achieving state-of-the-art performance in SMT. In this paper, we give a more systematic treatment by summarizing the relevant source information through a convolutional architecture guided by the target information. With different guiding signals during decoding, our specifically designed convolution+gating architectures can pinpoint the parts of a source sentence that are relevant to predicting a target word, and fuse them with the context of entire source sentence to form a unified representation. This representation, together with target language words, are fed to a deep neural network (DNN) to form a stronger NNJM. Experiments on two NIST Chinese-English translation tasks show that the proposed model can achieve significant improvements over the previous NNJM by up to +1.08 BLEU points on average
Cite
@article{arxiv.1503.01838,
title = {Encoding Source Language with Convolutional Neural Network for Machine Translation},
author = {Fandong Meng and Zhengdong Lu and Mingxuan Wang and Hang Li and Wenbin Jiang and Qun Liu},
journal= {arXiv preprint arXiv:1503.01838},
year = {2015}
}
Comments
Accepted as a full paper at ACL 2015