AgentSchool: An LLM-Powered Multi-Agent Simulation for Education
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2026
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917544530542592 |
|---|---|
| author | Ye, Yulei Li, Wenhao Wen, Zhong Huang, Yunshu Hu, Yichen Wei, Zifan Wang, Yige Xie, Xinyu Yang, Haoxuan Huang, Yanjun Li, Ruijia Qian, Hong Song, Yu Jiang, Bo Li, Bingdong Li, Lijun Zhang, Bo Cai, Pinlong Xu, Xingcheng Chen, Shuangye Hu, Xia He, Liang Zhou, Aimin Qu, Jingjing Shao, Jing Wang, Xiangfeng |
| author_facet | Ye, Yulei Li, Wenhao Wen, Zhong Huang, Yunshu Hu, Yichen Wei, Zifan Wang, Yige Xie, Xinyu Yang, Haoxuan Huang, Yanjun Li, Ruijia Qian, Hong Song, Yu Jiang, Bo Li, Bingdong Li, Lijun Zhang, Bo Cai, Pinlong Xu, Xingcheng Chen, Shuangye Hu, Xia He, Liang Zhou, Aimin Qu, Jingjing Shao, Jing Wang, Xiangfeng |
| contents | Despite the rapid deployment of LLMs into classrooms, validating educational AI remains uniquely intractable: interventions act on developing learners whose cognitive and social trajectories are irreversibly shaped, while real-world trials are slow, ethically constrained, and institutionally locked. LLM-based educational simulators have emerged as a potential remedy, but many still collapse learning into persona-conditioned role-play and, when optimized only to reproduce existing classrooms, can structurally penalize the institutional novelty that pedagogical reform requires. In this work, we introduce AgentSchool, an LLM-driven multi-agent simulator that models learning as state transition rather than prompted behavior. AgentSchool couples cognitively growable student agents -- equipped with weighted subject knowledge graphs, thinking-workflow pools, and explicit misconceptions -- with adaptive teacher agents that plan, scaffold, and reflect along the Zone of Proximal Development, embedded in a configurable scenery generator that situates instruction within both formal and informal learning fields, and a multi-scale simulator that decouples interaction scale, temporal granularity, and simulation duration. Experiments show that structured student agents produce more differentiated mastery and misconception traces than a baseline simulator, while teacher-agent comparisons show backbone-dependent patterns consistent with ZPD-informed adaptation. Further, AgentSchool generates plausible traces of peripheral participation, clique formation, aggressor-induced cohesion, and opinion-leader emergence consistent with classroom social theories. Beyond its role as an educational research instrument, AgentSchool frames education as a socially meaningful testbed for long-horizon memory, multi-agent coordination, and future institutional reasoning under organizational pressure. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_30144 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AgentSchool: An LLM-Powered Multi-Agent Simulation for Education Ye, Yulei Li, Wenhao Wen, Zhong Huang, Yunshu Hu, Yichen Wei, Zifan Wang, Yige Xie, Xinyu Yang, Haoxuan Huang, Yanjun Li, Ruijia Qian, Hong Song, Yu Jiang, Bo Li, Bingdong Li, Lijun Zhang, Bo Cai, Pinlong Xu, Xingcheng Chen, Shuangye Hu, Xia He, Liang Zhou, Aimin Qu, Jingjing Shao, Jing Wang, Xiangfeng Artificial Intelligence Multiagent Systems Despite the rapid deployment of LLMs into classrooms, validating educational AI remains uniquely intractable: interventions act on developing learners whose cognitive and social trajectories are irreversibly shaped, while real-world trials are slow, ethically constrained, and institutionally locked. LLM-based educational simulators have emerged as a potential remedy, but many still collapse learning into persona-conditioned role-play and, when optimized only to reproduce existing classrooms, can structurally penalize the institutional novelty that pedagogical reform requires. In this work, we introduce AgentSchool, an LLM-driven multi-agent simulator that models learning as state transition rather than prompted behavior. AgentSchool couples cognitively growable student agents -- equipped with weighted subject knowledge graphs, thinking-workflow pools, and explicit misconceptions -- with adaptive teacher agents that plan, scaffold, and reflect along the Zone of Proximal Development, embedded in a configurable scenery generator that situates instruction within both formal and informal learning fields, and a multi-scale simulator that decouples interaction scale, temporal granularity, and simulation duration. Experiments show that structured student agents produce more differentiated mastery and misconception traces than a baseline simulator, while teacher-agent comparisons show backbone-dependent patterns consistent with ZPD-informed adaptation. Further, AgentSchool generates plausible traces of peripheral participation, clique formation, aggressor-induced cohesion, and opinion-leader emergence consistent with classroom social theories. Beyond its role as an educational research instrument, AgentSchool frames education as a socially meaningful testbed for long-horizon memory, multi-agent coordination, and future institutional reasoning under organizational pressure. |
| title | AgentSchool: An LLM-Powered Multi-Agent Simulation for Education |
| topic | Artificial Intelligence Multiagent Systems |
| url | https://arxiv.org/abs/2605.30144 |