Graph-based Learning for Statistical Machine Translation

Current phrase-based statistical machine translation systems process each test sentence in isolation and do not enforce global consistency constraints, even though the test data is often internally consistent with respect to topic or style. We propose a new consistency model for machine translation in the form of a graph-based semi-supervised learning algorithm that exploits similarities between training and test data and also similarities between different test sentences. The algorithm learns a regression function jointly over training and test data and uses the resulting scores to rerank translation hypotheses. Evaluation on two travel expression translation tasks demonstrates improvements of up to 2.6 BLEU points absolute and 2.8% in PER.