Combinatory Categorial Grammar (CCG) is a lexicalized grammar formalism in which words are associated with categories that specify the syntactic configurations in which they may occur. We present a novel parsing model with the capacity to capture the associative adjacent-category relationships intrinsic to CCG by parameterizing the relationships between each constituent label and the preterminal categories directly to its left and right, biasing the model toward constituent categories that can combine with their contexts. This builds on the intuitions of Klein and Manning's (2002) "constituent-context" model, which demonstrated the value of modeling context, but has the advantage of being able to exploit the properties of CCG. Our experiments show that our model outperforms a baseline in which this context information is not captured.

PDF

In Proceedings of the 2015 Conference on Computational Natural Language Learning (CoNLL-2015), pp. 22--31, Beijing, China 2015.

@inproceedings{garrette:conll15, title={A Supertag-Context Model for Weakly-Supervised CCG Parser Learning}, author={Dan Garrette and Chris Dyer and Jason Baldridge and Noah A. Smith }, booktitle={Proceedings of the 2015 Conference on Computational Natural Language Learning (CoNLL-2015)}, address={Beijing, China}, pages={22--31}, url="http://www.cs.utexas.edu/users/ai-labpub-view.php?PubID=127522", year={2015} }

Slides (PDF)