We describe a method for recognizing textual entailment that uses the length of the longest common subsequence (LCS) between two texts as its decision criterion. Rather than requi...
Psycholinguistic studies suggest a model of human language processing that 1) performs incremental interpretation of spoken utterances or written text, 2) preserves ambiguity by m...
William Schuler, Samir AbdelRahman, Tim Miller, La...
This paper presents the development of an open-source Spanish Dependency Grammar implemented in FreeLing environment. This grammar was designed as a resource for NLP applications ...
We explore a new Bayesian model for probabilistic grammars, a family of distributions over discrete structures that includes hidden Markov models and probabilistic context-free gr...
Discriminative methods have shown significant improvements over traditional generative methods in many machine learning applications, but there has been difficulty in extending th...