Park et al. 2024

LLM agents grounded in self-reports enable general-purpose simulation of individuals. Park, J. S., et al. (2024, rev. 2026). arXiv:2411.10109v3.

Relationship to this project — Rich-conditioning contrast

Shows rich interview conditioning can substantially outperform lightweight persona descriptions. This paper studies the lightweight end deliberately: the cheap construction that practitioners actually deploy at scale.