Modeling Term Translation for Document-informed Machine Translation

Term translation is of great importance for statistical machine translation (SMT), es-pecially document-informed SMT. In this paper, we investigate three issues of term translation in the context of document-informed SMT and propose three cor-responding models: (a) a term trans-lation disambiguation model which se-lects desirable translations for terms in the source language with domain information, (b) a term translation consistency model that encourages consistent translations for terms with a high strength of translation consistency throughout a document, and (c) a term bracketing model that rewards translation hypotheses where bracketable source terms are translated as a whole unit. We integrate the three models into hierarchical phrase-based SMT and eval-uate their effectiveness on NIST Chinese-English translation tasks with large-scale training data. Experiment results show that all three models can achieve sig-nificant improvements over the baseline. Additionally, we can obtain a further improvement when combining the three models.