Unidirectional derivation semantics for synchronous tree-adjoining grammars

2Citations
Citations of this article
5Readers
Mendeley users who have this article in their library.
Get full text

Abstract

Synchronous tree-adjoining grammars have been given two types of semantics: one based on bimorphisms and one based on synchronous derivations, in both of which the input and output trees are constructed synchronously. We introduce a third type of semantics that is based on unidirectional derivations. It derives output trees based on a given input tree and thus marks a first step towards conditional probability distributions. We prove that the unidirectional semantics coincides with the bimorphism-based semantics with the help of a strong correspondence to linear and nondeleting extended top-down tree transducers with explicit substitution. In addition, we show that stateful synchronous tree-adjoining grammars admit a normal form in which only adjunction is used. This contrasts the situation encountered in the stateless case. © 2012 Springer-Verlag.

Cite

CITATION STYLE

APA

Büchse, M., Maletti, A., & Vogler, H. (2012). Unidirectional derivation semantics for synchronous tree-adjoining grammars. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 7410 LNCS, pp. 368–379). https://doi.org/10.1007/978-3-642-31653-1_33

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free