Ron Petrick's Webpages

Unpublished manuscript

Domain proposal: Sentence generation with tree-adjoining grammars (CRISP), A. Koller and R. Petrick, planning domain proposal originally submitted to the deterministic track of the 2011 International Planning Competition (IPC 2011), unpublished, 2010.

The CRISP domain was not used in IPC 2011, however, the problem generator, domain definition, and sample problems are available from the IPC 2011 unused domains page.

[ bib | pdf ]

Abstract

The CRISP domain is motivated by the problem of sentence generation in natural language generation (NLG). (The name CRISP refers to an NLG system by Koller and Stone (2007) that attempts to solve the sentence generation problem using planning techniques.) NLG is a major subfield of natural language processing, concerned with computing natural language sentences or texts that convey a given piece of information to an audience. In the sentence generation task, we focus on generating a single sentence that expresses a given meaning according to a given grammar. As a planning task, a plan encodes the necessary sentence with the actions in the plan corresponding to the utterance of individual words arising from a syntactically correct (and semantically appropriate) grammatical derivation.