Abstract

We present a variation on classic beam thresholding techniques that is up to an order of magnitude faster than the traditional method, at the same performance level. We also present a new thresholding technique, global thresholding, which, combined with the new beam thresholding, gives an additional factor of two improvement, and a novel technique, multiple pass parsing, that can be combined with the others to yield yet another 50% improvement. We use a new search algorithm to simultaneously op- timize the thresholding parameters of the various algorithms.

Citations

396	New Statistical Parser Based on Bigram Lexical De-pendencies - Collins - 2006
203	Tree-Bank Grammars - Charniak - 1996
155	An e cient probabilistic context-free parsing algorithm that computes pre x probabilities - Stolcke - 1995
112	Stochastic lexicalized tree-adjoining grammars - Schabes - 1992
83	Probabilistic tree-adjoining grammar as a framework for statistical natural language processing - Resnik - 1992
79	Grammatical trigrams: a probabilistic model of link grammar - Lafferty, Sleator, et al. - 1992
69	Tree Insertion Grammar: A Cubic-Time Parsable Formalism that Lexicalizes ContextFree Grammar without Changing the Tree Produced. Technical Report TR-94-13, Mitsubishi Electric Research Laboratories. A Parameters of PLTIG Each elementary tree of a Probabil - Schabes, Waters - 1994
68	A polynomial-time algorithm for statistical machine translation - Wu - 1996
65	New figures of merit for best first probabilistic chart parsing - Caraballo, Charniak - 1997
22	Fast parsing using pruning and grammar specialization - Rayner, Carter - 1996
4	New uses for the n-best sentence hypothesis within the Byblos speech recognition system - Schwartz, Austin, et al. - 1992