Minimum-error-rate training (MERT) is a bottleneck for current development in statistical machine translation because it is limited in the number of weights it can reliably optimi...
Kernel-based learning (e.g., Support Vector Machines) has been successfully applied to many hard problems in Natural Language Processing (NLP). In NLP, although feature combinatio...
Constraint networks have been shown to be useful in formulating such diverse problems as scene labeling, natural language parsing, and temporal reasoning. Given a constraint netwo...
In this paper we present a framework for semantic scene parsing and object recognition based on dense depth maps. Five viewindependent 3D features that vary with object class are e...
Toponym Disambiguation, i.e. the task of assigning to place name their correct reference in the world, is getting more attention from many researchers. Many methods have been prop...