Translation Model Generalization using Probability Averaging for Machine Translation

Previous methods on improving translation quality by employing multiple SMT models usually carry out as a secondpass decision procedure on hypotheses from multiple systems using extra features instead of using features in existing models in more depth. In this paper, we propose translation model generalization (TMG), an approach that updates probability feature values for the translation model being used based on the model itself and a set of auxiliary models, aiming to enhance translation quality in the firstpass decoding. We validate our approach on translation models based on auxiliary models built by two different ways. We also introduce novel probability variance features into the log-linear models for further improvements. We conclude that our approach can be developed independently and integrated into current SMT pipeline directly. We demonstrate BLEU improvements on the NIST Chinese-toEnglish MT tasks for single-system decodings, a system combination approach and a model combination approach. 1