Improving Statistical Machine Translation Performance by Oracle-BLEU Model Re-estimation

We present a novel technique for training translation models for statistical machine translation by aligning source sentences to their oracle-BLEU translations. In contrast to previous approaches which are constrained to phrase training, our method also allows the re-estimation of reordering models along with the translation model. Experiments show an improvement of up to 0.8 BLEU for our approach over a competitive Arabic-English baseline trained directly on the word-aligned bitext using heuristic extraction. As an additional benefit, the phrase table size is reduced dramatically to only 3% of the original size.