Tag
This paper introduces a novel paradigm, VLM-as-probabilistic-grounder, which models uncertainty in vision-language model groundings as probability distributions for symbolic belief-space planning, enhancing robustness in partially observable settings.
This paper introduces YUKTI, a framework that transforms natural-language decision situations into robust, verifiable decisions by using an uncertainty-typed proposition graph, assumption-robust Pareto frontiers with regret bounds, and a multi-stage optimization hand-off. Validated on synthetic and real datasets, it significantly reduces regret compared to naive point-solution approaches and identifies the limits of language model reasoning.