Vector representations of generative models and their consistent estimation

Music City Center 

Generative models, like large language models or text-to-image diffusion models, can generate a random output or response after being given a query from a user. Representing them with vectors in a finite-dimensional Euclidean space based on their responses to a set of queries, facilitates statistical decision-making tasks on black-box generative models using conventional tools. We establish sufficient conditions for consistent estimation of population-level vector representations of a set of generative models based on their sample responses to a set of queries.

generative models

multidimensional scaling

raw stress embedding 

Section on Statistical Learning and Data Science