Paper: A Probabilistic Model of Referring Expressions for Complex Objects

ACL ID W09-0634
Title A Probabilistic Model of Referring Expressions for Complex Objects
Venue European Workshop on Natural Language Generation
Session
Year 2009
Authors

This paper presents a probabilistic model both for generation and understanding of referring expressions. This model intro- ducestheconceptofparts of objects, mod- elling the necessity to deal with the char- acteristics of separate parts of an object in the referring process. This was ignored or implicit in previous literature. Integrating this concept into a probabilistic formula- tion, the model captures human character- istics of visual perception and some type of pragmatic implicature in referring ex- pressions. Developing this kind of model is critical to deal with more complex do- mains in the future. As a first step in our research, we validate the model with the TUNA corpus to show that it includes con- ventional domain modeling as a subset.