Abstract

We present a statistical phrase-based translation model that uses hierarchical phrases— phrases that contain subphrases. The model is formally a synchronous context-free grammar but is learned from a bitext without any syntactic information. Thus it can be seen as a shift to the formal machinery of syntaxbased translation systems without any linguistic commitment. In our experiments using BLEU as a metric, the hierarchical phrasebased model achieves a relative improvement of 7.5 % over Pharaoh, a state-of-the-art phrase-based system. 1

Citations

992	BLEU: A Method for Automatic Evaluation of Machine Translation - Papineni, Roukos, et al. - 2002
892	Mathematics of Statistical Machine Translation: Parameter Estimation - Brown, Pietra, et al. - 1993
631	J: An Empirical Study of Smoothing Techniques for Language Modeling - SF, Goodman - 1996
343	Stochastic inversion transduction grammars and bilingual parsing of parallel corpora - Wu - 1997
341	Improved statistical alignment models - Och, Ney - 2000
256	Discriminative training and maximum entropy models for statistical machine translation - Och, Ney
245	The alignment template approach to statistical machine translation - Och, Ney - 2004
202	K.: A syntax-based statistical translation model - Yamada, Knight - 2001
58	Syntax directed translations and the pushdown assembler - Aho, Ullman - 1969
44	Improvements in phrase-based statistical machine translation - Zens, Ney - 2004
11	Pharaoh: A beam search decoder for phrase-based statistical machine translation models - 2004a
11	Statistical significance tests for machine translation evaluation - 2004b
7	Example-Based Incremental Synchronous Interpretation - Block - 2000