Prompt Optimization Across Multiple Agents for Representing Diverse Human Populations
This addresses the need for scalable and diverse proxies for human behavior in applications like crowdsourcing and education, though it is incremental as it builds on prior work on LLM diversity and submodular optimization.
The paper tackles the problem of LLMs producing homogeneous outputs that fail to capture human diversity by proposing a framework to construct a set of LLM agents that collectively represent diverse human populations, with experiments showing they more effectively represent human populations compared to baselines.
The difficulty and expense of obtaining large-scale human responses make Large Language Models (LLMs) an attractive alternative and a promising proxy for human behavior. However, prior work shows that LLMs often produce homogeneous outputs that fail to capture the rich diversity of human perspectives and behaviors. Thus, rather than trying to capture this diversity with a single LLM agent, we propose a novel framework to construct a set of agents that collectively capture the diversity of a given human population. Each agent is an LLM whose behavior is steered by conditioning on a small set of human demonstrations (task-response pairs) through in-context learning. The central challenge is therefore to select a representative set of LLM agents from the exponentially large space of possible agents. We tackle this selection problem from the lens of submodular optimization. In particular, we develop methods that offer different trade-offs regarding time complexity and performance guarantees. Extensive experiments in crowdsourcing and educational domains demonstrate that our approach constructs agents that more effectively represent human populations compared to baselines. Moreover, behavioral analyses on new tasks show that these agents reproduce the behavior patterns and perspectives of the students and annotators they are designed to represent.