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

arXiv: https://arxiv.org/abs/2602.14279 · HTML: https://mlsanity.com/paper/2602.14279/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 15 of its 33 references that are in the map.
- First posted on arXiv 15 Feb 2026.

## Builds on (most cited first, 15 of 15)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [MetaGPT: Meta Programming for A Multi-Agent Collaborative Framework](https://mlsanity.com/paper/2308.00352/) (2023)
- [Modeling Relational Data with Graph Convolutional Networks](https://mlsanity.com/paper/1703.06103/) (2017)
- [Relational inductive biases, deep learning, and graph networks](https://mlsanity.com/paper/1806.01261/) (2018)
- [An Explanation of In-context Learning as Implicit Bayesian Inference](https://mlsanity.com/paper/2111.02080/) (2021)
- [Whose Opinions Do Language Models Reflect?](https://mlsanity.com/paper/2303.17548/) (2023)
- [MedAgents: Large Language Models as Collaborators for Zero-shot Medical Reasoning](https://mlsanity.com/paper/2311.10537/) (2023)
- [Debating with More Persuasive LLMs Leads to More Truthful Answers](https://mlsanity.com/paper/2402.06782/) (2024)
- [Modern Bayesian Experimental Design](https://mlsanity.com/paper/2302.14545/) (2023)
- [Deep Adaptive Design: Amortizing Sequential Bayesian Experimental Design](https://mlsanity.com/paper/2103.02438/) (2021)
- [Implicit Deep Adaptive Design: Policy-Based Experimental Design without Likelihoods](https://mlsanity.com/paper/2111.02329/) (2021)
- [Martingale posterior distributions](https://mlsanity.com/paper/2103.15671/) (2021)
- [Twin-2K-500: A dataset for building digital twins of over 2,000 people based on their answers to over 500 questions](https://mlsanity.com/paper/2505.17479/) (2025)
- [Adaptive Elicitation of Latent Information Using Natural Language](https://mlsanity.com/paper/2504.04204/) (2025)
- [Graph-Based Alternatives to LLMs for Human Simulation](https://mlsanity.com/paper/2511.02135/) (2025)

## Abstract (arXiv)

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.