Whom to Query for What: Adaptive Group Elicitation via Multi-Turn LLM Interactions

TL;DR

This paper introduces an adaptive multi-turn elicitation framework using large language models and graph neural networks to efficiently select questions and respondents, improving population response prediction under budget constraints.

Contribution

It presents a novel, theoretically grounded approach combining LLM-based information gain and graph neural networks for adaptive group elicitation with partial responses.

Findings

Achieves over 12% relative gain in population response prediction on CES dataset.

Effectively selects small, informative respondent subsets under budget constraints.

Improves response prediction accuracy across three real-world opinion datasets.

Abstract

Eliciting information to reduce uncertainty about latent group-level properties from surveys and other collective assessments requires allocating limited questioning effort under real costs and missing data. Although large language models enable adaptive, multi-turn interactions in natural language, most existing elicitation methods optimize what to ask with a fixed respondent pool, and do not adapt respondent selection or leverage population structure when responses are partial or incomplete. To address this gap, we study adaptive group elicitation, a multi-round setting where an agent adaptively selects both questions and respondents under explicit query and participation budgets. We propose a theoretically grounded framework that combines (i) an LLM-based expected information gain objective for scoring candidate questions with (ii) heterogeneous graph neural network propagation that aggregates observed responses and participant attributes to impute missing responses and guide per-round respondent selection. This closed-loop procedure queries a small, informative subset of individuals while inferring population-level responses via structured similarity. Across three real-world opinion datasets, our method consistently improves population-level response prediction under constrained budgets, including a >12% relative gain on CES at a 10% respondent budget.