Search and Filter

Eduverse: A User-Defined Multi-Agent Simulation Space For Education Scenario

Date

AI Overview

Why it matters

Educational AI systems mostly focus on short-term interactions or single agents, failing to capture evolving student cognition, social group dynamics, and multi-session development in complex virtual classrooms.

Study at a glance
  • Question: Can customizable multi-agent simulations realistically reproduce classroom interactions, group dynamics, and student learning over time?
  • Setting and data: Simulated junior secondary Chinese language classrooms, six virtual student archetypes, one teacher agent, three text genres across four consecutive sessions.
  • Method: Multi-agent system development and simulation evaluation with ablation studies and human-in-the-loop testing.
  • AI tools: EduVerse, GPT-4, InternVL
Key findings
  • Simulated teacher-student dialogue completion rates ranged from 0.28 to 0.64, closely matching real classroom rates of 0.37 to 0.49 across three text genres.
  • Positive developmental shifts in student behavior, emotion, and cognition increased by an average of 11.7% across four consecutive sessions, with behavior improving most rapidly.
  • Interaction networks realized 27% to 40% of potential peer connections across classroom layouts, while human participants successfully integrated into virtual discussions with personality-consistent student agents.

Summary drafted with AI from the full paper and may contain errors. See the full paper for details.

Reproducing cognitive development, group interaction, and long-term evolution in virtual classrooms remains a core challenge for educational AI, as real classrooms integrate open-ended cognition, dynamic social interaction, affective factors, and multi-session development rarely captured together. Existing approaches mostly focus on short-term or single-agent settings, limiting systematic study of classroom complexity and cross-task reuse. We present EduVerse, the first user-defined multi-agent simulation space that supports environment, agent, and session customization. A distinctive human-in-the-loop interface further allows real users to join the space. Built on a layered CIE (Cognition-Interaction-Evolution) architecture, EduVerse ensures individual consistency, authentic interaction, and longitudinal adaptation in cognition, emotion, and behavior-reproducing realistic classroom dynamics with seamless human-agent integration. We validate EduVerse in middle-school Chinese classes across three text genres, environments, and multiple sessions. Results show: (1) Instructional alignment: simulated IRF rates (0.28-0.64) closely match real classrooms (0.37-0.49), indicating pedagogical realism; (2) Group interaction and role differentiation: network density (0.27-0.40) with about one-third of peer links realized, while human-agent tasks indicate a balance between individual variability and instructional stability; (3) Cross-session evolution: the positive transition rate R+ increase by 11.7% on average, capturing longitudinal shifts in behavior, emotion, and cognition and revealing structured learning trajectories. Overall, EduVerse balances realism, reproducibility, and interpretability, providing a scalable platform for educational AI. The system will be open-sourced to foster cross-disciplinary research.