In this paper, we derive a notion of 'word meaning in context' which accounts for the wide range of lexical shifts and ambiguities observed in utterance interpretation. We characterize the lexical comprehension process as a combination of cognitive semantics and Discourse Representation Theory, formalized as a 'situation description system': a probabilistic model which takes utterance understanding to be the mental process of describing one or more situations that would account for an observed utterance. Our model uses insights from different types of generative models to capture the interplay of local and global contexts and their joint influence upon the lexical representation of sentence constituents. We implement the system using a directed graphical model, and apply it to examples containing various contextualisation phenomena.
View on arXiv