C04-1030 begin with the IBM constraints for phrase-based translation . Then , we will describe constraints
D09-1107 propose is also related to standard phrase-based translation models through the use of the
C04-1030 Tillmann and Ney , 2003 ) . For the phrase-based translation approach , we use the same idea
D08-1033 translation systems . 1 Introduction In phrase-based translation , statistical knowledge of translation
D09-1021 algorithms originally developed for phrase-based translation systems . Experiments on translation
C04-1030 investigate the reordering problem for phrase-based translation approaches . As the word order
D08-1060 effectively enhances the quality of the phrase-based translation of Chinese-to-English . One major
D09-1021 describes how phrase entries from phrase-based translation systems can be modified to include
D09-1105 both training and evaluation of a phrase-based translation system , namely Moses . Our methods
C04-1030 a joint probability model for phrase-based translation . In ( Koehn et for the SLDB
D10-1014 feature into the hierarchical phrase-based translation system also brings in additional
D10-1014 the advantages of hierarchical phrase-based translation systems while at the same time
D10-1014 the advantages of hierarchical phrase-based translation systems while at the same time
D09-1105 ? The baseline system is Moses phrase-based translation with no preprocessing and only
C04-1030 % . 5 Related Work Recently , phrase-based translation approaches became more and more
D09-1105 translation in the target language . Phrase-based translation systems rely heavily on the target
D09-1123 Roukos , 2007 ) , expresses the phrase-based translation task in a unified log-linear
D08-1033 Alignment Model While state-of-the-art phrase-based translation systems include an increasing
D09-1105 using different versions of a phrase-based translation system . Many others have proposed
D10-1014 syntactic constraints . 2 Hierarchical Phrase-Based Translation An SCFG is a synchronous rewriting
hide detail