Paper: Predicting the Resolution of Referring Expressions from User Behavior

ACL ID D13-1134
Title Predicting the Resolution of Referring Expressions from User Behavior
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2013
Authors

We present a statistical model for predicting how the user of an interactive, situated NLP system resolved a referring expression. The model makes an initial prediction based on the meaning of the utterance, and revises it con- tinuously based on the user?s behavior. The combined model outperforms its components in predicting reference resolution and when to give feedback.