Large-scale Exact Decoding: The IMS-TTT submission to WMT14
Daniel Quernheim, Fabienne Cap · 2014
We present the IMS-TTT submission to WMT14, an experimental statistical treeto-tree machine translation system based on the multi-bottom up tree transducer including rule extraction, tuning and decoding.Thanks to input parse forests and a "no pruning" strategy during decoding, the obtained translations are competitive.The drawbacks are a restricted coverage of 70% on test data, in part due to exact input parse tree matching, and a relatively high runtime.Advantages include easy redecoding with a different weight vector, since the full translation forests can be stored after the first decoding pass. *This work was supported by Deutsche Forschungsgemeinschaft grants Models of Morphosyntax for Statistical Machine Translation (Phase 2) and MA/4959/1-1.1 A translation is sensible if it is of linear size increase and can be computed by some (potentially copying) top-down tree transducer.