arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2605.30144cs.AIcs.MA
收藏

AgentSchool:基于LLM的多智能体教育模拟系统

AgentSchool: An LLM-Powered Multi-Agent Simulation for Education

Shanghai Institute of AI for Education(上海人工智能教育研究院);School of Computer Science and Technology(计算机科学与技术学院);East China Normal University(东华大学);School of Design(设计学院);Faculty of Education(教育学院);Shanghai Artificial Intelligence Laboratory(上海人工智能实验室)
Yulei Ye, Wenhao Li, Zhong Wen, Yunshu Huang, Yichen Hu, Zifan Wei, Yige Wang, Xinyu Xie, Haoxuan Yang, Yanjun Huang, Ruijia Li, Hong Qian, Yu Song, Bo Jiang, B… 展开作者

Yulei Ye, Wenhao Li, Zhong Wen, Yunshu Huang, Yichen Hu, Zifan Wei, Yige Wang, Xinyu Xie, Haoxuan Yang, Yanjun Huang, Ruijia Li, Hong Qian, Yu Song, Bo Jiang, Bingdong Li, Lijun Li, Bo Zhang, Pinlong Cai, Xingcheng Xu, Shuangye Chen, Xia Hu, Liang He, Aimin Zhou, Jingjing Qu, Jing Shao, Xiangfeng Wang

AI总结:

提出AgentSchool,一种LLM驱动的多智能体模拟器,通过可成长的学生智能体(带知识图谱、思维工作流和错误概念)与自适应教师智能体(基于最近发展区)模拟学习过程,支持多尺度模拟,实验验证了其生成差异化掌握轨迹和符合课堂社会理论的行为模式。

摘要:

尽管LLM已迅速部署到课堂中,验证教育AI仍然具有独特的棘手性:干预措施作用于发展中的学习者,其认知和社会轨迹被不可逆地塑造,而现实世界试验缓慢、受伦理约束且受制度限制。基于LLM的教育模拟器已成为潜在的补救措施,但许多模拟器仍将学习简化为角色扮演,并且当仅优化以再现现有课堂时,可能会结构性惩罚教学改革所需的制度创新。在这项工作中,我们介绍了AgentSchool,一种LLM驱动的多智能体模拟器,将学习建模为状态转换而非提示行为。AgentSchool将可成长的学生智能体(配备加权学科知识图谱、思维工作流池和显式错误概念)与自适应教师智能体(在最近发展区内规划、搭建支架和反思)相结合,嵌入可配置的场景生成器(将教学置于正式和非正式学习领域)和多尺度模拟器(解耦交互规模、时间粒度和模拟持续时间)。实验表明,结构化学生智能体比基线模拟器产生更差异化的掌握和错误概念轨迹,而教师智能体比较显示出与基于ZPD的适应一致的骨干依赖模式。此外,AgentSchool生成与课堂社会理论一致的外围参与、小团体形成、攻击者诱导的凝聚力和意见领袖出现的合理轨迹。除了作为教育研究工具的作用外,AgentSchool还将教育构建为在组织压力下进行长时记忆、多智能体协调和未来制度推理的社会意义测试平台。

英文摘要:

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.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑