Left-to-Right Target Generation for Hierarchical Phrase-Based Translation

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 777–784, Sydney, July 2006. c⃝2006 Association for Computational Linguistics Left-to-Right Target Generation for Hierarchical Phrase-based Translation Taro Watanabe Hajime Tsukada Hideki Isozaki 2-4, Hikaridai, Seika-cho, Soraku-gun, Kyoto, JAPAN 619-0237 {taro,tsukada,isozaki}@cslab.kecl.ntt.co.jp Abstract We present a hierarchical phrase-based statistical machine translation in which a target sentence is efficiently generated in left-to-right order. The model is a class of synchronous-CFG with a Greibach Nor- mal Form-like structure for the projected production rule: The paired target-side of a production rule takes a phrase pre- fixed form. The decoder for the target- normalized form is based on an Early- style top down parser on the source side. The target-normalized form coupled with our top down parser implies a left-to- right generation of translations which en- ables us a straightforward integration with ngram language models. Our model was experimented on a Japanese-to-English newswire translation task, and showed sta- tistically significant performance improve- ments against a phrase-based translation system. 1 Introduction In a classical statistical machine translation, a for- eign language sentence f J 1 = f1, f2, ... fJ is trans- lated into another language, i.e. English, eI 1 = e1, e2, ..., eI by seeking a maximum likely solution of: ˆeI 1 = argmax eI 1 Pr(eI 1| f J 1 ) (1) = argmax eI 1 Pr( f J 1 |eI 1)Pr(eI 1) (2) The source channel approach in Equation 2 inde- pendently decomposes translation knowledge into a translation model and a language model, respec- tively (Brown et al., 1993). The former repre- sents the correspondence between two languages and the latter contributes to the fluency of English. In the state of the art statistical machine transla- tion, the posterior probability Pr(eI 1| f J 1 ) is directly maximized using a log-linear combination of fea- ture functions (Och and Ney, 2002): ˆeI 1 = argmax eI 1 exp ��M m=1 λmhm(eI 1, f J 1 ) � � e′I′ 1 exp ��M m=1 λmhm(e′I′ 1 , f J 1 ) � (3) where hm(eI 1, f J 1 ) is a feature function, such as a ngram language model or a translation model. When decoding, the denominator is dropped since it depends only on f J 1 . Feature function scaling factors λm are optimized based on a maximum likely approach (Och and Ney, 2002) or on a direct error minimization approach (Och, 2003). This modeling allows the integration of various fea- ture functions depending on the scenario of how a translation is constituted. A phrase-based translation model is one of the modern approaches which exploits a phrase, a contiguous sequence of words, as a unit of transla- tion (Koehn et al., 2003; Zens and Ney, 2003; Till- man, 2004). The idea is based on a word-based source channel modeling of Brown et al. (1993): It assumes that eI of K phrases ¯eK1 is segmented into a sequence 1 . Each phrase ¯ek is transformed into …

Content truncated to 500 words