Abstract

Model generation calculi, close relatives of tableau calculi for theorem proving, can be used as competence models for semantic natural language understanding. Unfortunately, existing model generation calculi are not yet plausible as performance models of actual human processing, since they fail to capture computational aspects of human language processing.We outline an extended model generation calculus that solves the most unpleasant computational inadequacy; In the extended calculus, tableau expansion rules are equipped with costs, and model construction is a process that optimises model quality under resource constraints with respect to these costs. We embed the new calculus into an abstract inference machine and illustrate the possibilities of this approach by presenting a partial theory of definite descriptions in this setting.In this case study, the constants in the universe are given saliences, that are maintained across the model generation process. This additional data serves as one important source of information for model quality and resource cost estimation