Using Hypothesis Selection Based Features for Confusion Network MT System Combination
Sahar Ghannay, Loïc Barrault · 2014
This paper describes the development operated into MANY, an open source system combination software based on confusion networks developed at LIUM. The hypotheses from Chinese-English MT systems were combined with a new version of the software. MANY has been updated in order to use word confidence score and to boostn-grams occurring in input hypotheses. In this paper we propose either to use an adapted language model or adding some additional features in the decoder to boost certain n-grams probabilities. Experimental results show that the updates yielded significant improvements in terms of BLEU score.