Adaptive Querying with AI Persona Priors

2026-05-01

AI Summary

This work addresses the challenge of efficiently inferring user responses or psychometric traits for unseen items under a limited query budget, particularly in heterogeneous, high-dimensional, and cold-start settings. The authors propose an adaptive querying method grounded in AI-character priors: user states are modeled as mixture memberships over a finite dictionary of AI-generated personas, each endowed with response distributions synthesized by large language models. By integrating this representation with Bayesian sequential experimental design, the approach enables efficient item selection. The induced latent-variable model retains strong expressive capacity while permitting closed-form posterior updates and interpretable predictions, thereby circumventing restrictive parametric assumptions common in traditional methods. Empirical evaluations on both synthetic data and the WorldValuesBench demonstrate accurate probabilistic forecasting and yield transparent, adaptive questioning strategies.

Abstract

We study adaptive querying for learning user-dependent quantities of interest, such as responses to held-out items and psychometric indicators, within tight question budgets. Classical Bayesian design and computerized adaptive testing typically rely on restrictive parametric assumptions or expensive posterior approximations, limiting their use in heterogeneous, high-dimensional, and cold-start settings. We introduce a persona-induced latent variable model that represents a user's state through membership in a finite dictionary of AI personas, each offering response distributions produced by a large language model. This yields expressive priors with closed-form posterior updates and efficient finite-mixture predictions, enabling scalable Bayesian design for sequential item selection. Experiments on synthetic data and WorldValuesBench demonstrate that persona-based posteriors deliver accurate probabilistic predictions and an interpretable adaptive elicitation pipeline.

Problem

• adaptive querying
• user-dependent quantities
• cold-start
• high-dimensional
• heterogeneous

Innovation

• adaptive querying
• AI personas
• latent variable model
• Bayesian experimental design
• large language models