Weakly-Supervised Grammar-Informed Bayesian CCG Parser Learning

Combinatory Categorial Grammar (CCG) is a lexical-ized grammar formalism in which words are associated with categories that, in combination with a small uni-versal set of rules, specify the syntactic configurations in which they may occur. Previous work has shown that learning sequence models for CCG tagging can be im-proved by using priors that are sensitive to the formal properties of CCG as well as cross-linguistic univer-sals. We extend this approach to the task of learning a full CCG parser from weak supervision. We present a Bayesian formulation for CCG parser induction that assumes only supervision in the form of an incomplete tag dictionary mapping some word types to sets of po-tential categories. Our approach outperforms a base-line model trained with uniform priors by exploiting universal, intrinsic properties of the CCG formalism to bias the model toward simpler, more cross-linguistically common categories