Edu-Theater: A Data-Efficient Agent Framework for Scalable Learner Behavior Simulation through Staging Roll-Call

Edu-Theater is a newly proposed framework that utilizes a cohort-aware roll-call simulation paradigm to efficiently simulate learner behavior without the need for extensive individual interaction data. By leveraging a teacher agent and retrospective roll-call probing, Edu-Theater refines individual learner states from cohort-level proficiency priors, significantly reducing the number of required LLM calls while enhancing simulation accuracy. This approach is particularly beneficial for practitioners in educational AI, as it enables scalable learner simulations that improve adaptive testing and other applications with reduced data collection costs.