Paper: Feature Weight Optimization for Discourse-Level SMT

ACL ID W13-3308
Title Feature Weight Optimization for Discourse-Level SMT
Venue Discourse in Machine Translation
Year 2013

We present an approach to feature weight optimization for document-level decoding. This is an essential task for enabling future development of discourse-level statistical machine translation, as it allows easy inte- gration of discourse features in the decod- ing process. We extend the framework of sentence-level feature weight optimization to the document-level. We show experi- mentally that we can get competitive and relatively stable results when using a stan- dard set of features, and that this frame- work also allows us to optimize document- level features, which can be used to model discourse phenomena.