MaxSD: A Neural Machine Translation Evaluation Metric Optimized by Maximizing Similarity Distance

摘要：

　　We propose a novel metric for machine translation evaluation based on neural networks.In the training phrase,we maximize the distance between the similarity scores of high and low-quality hypotheses.Then,the trained neural network is used to evaluate the new hypotheses in the testing phase.The proposed metric can efficiently in-corporate lexical and syntactic metrics as features in the network and thus is able to capture different levels of linguistic information.Experiments on WMT-14 show state-of-the-art performance is achieved in two out of five language pairs on the system-level and one on the segmentlevel.Comparative results are also achieved in the remaining language pairs.

关键词： machine translation evaluation neural networks similarity distance maximization

作者: Qingsong Ma Fandong Meng Daqi Zheng Mingxuan Wang Yvette Graham Wenbin Jiang Qun Liu

作者单位: Key Laboratory of Intelligent Information Processing,Institute of Computing Technology,Chinese Acade ADAPT Centre,School of Computing,Dublin City University,Ireland Key Laboratory of Intelligent Information Processing,Institute of Computing Technology,Chinese Acade

会议类型: 国际会议

会议名称: 第五届自然语言处理与中文计算会议(NLPCC-ICCPOL2016)

会议地点: 昆明

会议语种:英文

页码: 1-9

在线出版日期: 2016-12-02（万方平台首次上网日期，不代表论文的发表时间）