Optimizing Automatic Evaluation of Machine Translation with the ListMLE Approach
Optimizing Automatic Evaluation of Machine Translation with the ListMLE Approach
复制标题
使用 ListMLE 方法优化机器翻译的自动评估
DOI:
10.1145/3226045
复制
发表时间:
2018-11
影响因子:
2
通讯作者:
王明文
中科院分区:
文献类型:
--
作者:
李茂西;王明文
Automatic evaluation of machine translation is critical for the evaluation and development of machine translation systems. In this study, we propose a new model for automatic evaluation of machine translation. The proposed model combines standard n-gram precision features and sentence semantic mapping features with neural features, including neural language model probabilities and the embedding distances between translation outputs and their reference translations. We optimize the model with a representative list-wise learning to rank approach, ListMLE, in terms of human ranking assessments. The experimental results on WMT’2015 Metrics task indicated that the proposed approach yields significantly better correlations with human assessments than several state-of-the-art baseline approaches. In particular, the results confirmed that the proposed list-wise learning to rank approach is useful and powerful for optimizing automatic evaluation metrics in terms of human ranking assessments. Deep analysis also demonstrated that optimizing automatic metrics with the ListMLE approach is a reasonable method and adding the neural features can gain considerable improvements compared with the traditional features.
登录
查看更多内容
DOI:
--
发表时间:
2010
期刊:
--
影响因子:
--
作者:
Lucia Specia;J. Giménez
通讯作者:
Lucia Specia;J. Giménez
DOI:
10.1145/2815619
发表时间:
2016-01
影响因子:
2
作者:
Maoxi Li;Mingwen Wang;Hanxi Li;Fan Xu
通讯作者:
Fan Xu
DOI:
10.18653/v1/w15-3047
发表时间:
2015-09
期刊:
--
影响因子:
--
作者:
Rohit Gupta;Constantin Orasan;Josef van Genabith
通讯作者:
Rohit Gupta;Constantin Orasan;Josef van Genabith
DOI:
--
发表时间:
2005
期刊:
--
影响因子:
--
作者:
M. Snover;B. Dorr;R. Schwartz;Linnea Micciulla;R. Weischedel
通讯作者:
M. Snover;B. Dorr;R. Schwartz;Linnea Micciulla;R. Weischedel
DOI:
--
发表时间:
2010-07
期刊:
--
影响因子:
--
作者:
Chang Liu;Daniel Dahlmeier;H. Ng
通讯作者:
Chang Liu;Daniel Dahlmeier;H. Ng