Large Language Models That Replace Human Participants Can Harmfully Misportray and Flatten Identity Groups post
Unmasking Implicit Bias: Evaluating Persona-prompted LlM Responses in Power-Disparate Social Scenarios post