AI RESEARCH

Surprisal Minimisation over Goal-directed Alternatives Predicts Production Choice in Dialogue

arXiv CS.CL

ArXi:2605.00506v1 Announce Type: new We model utterance production as probabilistic cost-sensitive choice over contextual alternatives, using information-theoretic notions of cost. We distinguish between goal-directed alternatives that realise a fixed communicative intent and goal-agnostic alternatives defined only by contextual plausibility, allowing us to derive speaker- and listener-oriented interpretations of different cost measures. We present a procedure to generate both types of alternative sets using language models.