Improving the Performance of GIZA++ Using Variational Bayes

Bayesian approaches have been shown to reduce the amount of overfitting that occurs when running the EM algorithm, by placing prior probabilities on the model parameters. We apply one such Bayesian technique, variational Bayes, to GIZA++, a widely-used piece of software that computes word alignments for statistical machine translation. We show that using variational Bayes improves the performance of GIZA++, as well as improving the overall performance of the Moses machine translation system in terms of BLEU score. This work was supported by NSF grants IIS-0546554 and IIS-0910611.