Minimum Bayes-Risk Decoding for Statistical Machine Translation

We present Minimum Bayes-Risk (MBR) decoding for statistical machine translation. This statistical approach aims to minimize expected loss of translation errors under loss functions that measure translation performance. We describe a hierarchy of loss functions that incorporate different levels of linguistic information from word strings, word-to-word alignments from an MT system, and syntactic structure from parse-trees of source and target language sentences. We report the performance of the MBR decoders on a Chinese-to-English translation task. Our results show that MBR decoding can be used to tune statistical MT performance for specific loss functions.

• Type: Conference
• Year: 2004
• Where: NAACL
• Authors: Shankar Kumar, William J. Byrne
• Keywords: Loss Functions, NAACL 2004, NAACL 2007, Specific Loss Functions, Statistical Machine Translation