Joint Decoding with Multiple Translation Models

Current SMT systems usually decode with single translation models and cannot benefit from the strengths of other models in decoding phase. We instead propose joint decoding, a method that combines multiple translation models in one decoder. Our joint decoder draws connections among multiple models by integrating the translation hypergraphs they produce individually. Therefore, one model can share translations and even derivations with other models. Comparable to the state-of-the-art system combination technique, joint decoding achieves an absolute improvement

• Type: Journal
• Year: 2009
• Where: ACL
• Authors: Yang Liu, Haitao Mi, Yang Feng, Qun Liu
• Keywords: ACL 2009, Computational Linguistics, Joint, Single Translation Models, Translation Models