Teacher-agent comparisons show backbone-dependent patterns consistent with ZPD-informed adaptation¶

Claim · All claims
Evidence · 1 study · design r2 · q2
Currency · newest study 2026 · not yet checked against its sources · not reviewed by a person · recheck by 2029: tool and model accuracy changes quickly

Subclaims¶

q2 i? Teacher-agent comparison patterns vary by backbone LLM and are consistent with ZPD-informed adaptation. → Yulei Ye 2026

Evidence¶

Yulei Ye 2026¶

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. (2026). AgentSchool: An LLM-Powered Multi-Agent Simulation for Education. arXiv. https://arxiv.org/abs/2605.30144 doi:10.48550/arxiv.2605.30144

q2 · i? · design · r2

Teacher-agent comparisons from the same preliminary lesson study across five backbone LLMs. The paper reports "backbone-dependent patterns consistent with ZPD-informed adaptation"; no effect sizes or test statistics are printed in the available text.

"while teacher-agent comparisons show backbone-dependent patterns consistent with ZPD-informed adaptation"

Discussion¶

Related Claims¶

• In informal social scenes, AgentSchool generates plausible traces of peripheral participation, clique formation, aggressor-induced cohesion, and opinion-leader emergence consistent with classroom social theories — related
• In a preliminary 2×3 controlled lesson study across five backbone LLMs, structured student agents produce more differentiated mastery and misconception traces than a baseline simulator — related
• Expert evaluation finds larger, more capable LLM backbones produce higher-quality multi-domain graphs, with GPT-5 pro rated best — related
• In pairwise LLM-judge comparisons PsyScore-AEF wins over 90% in Actionability against GPT-4o and Llama, and removing the psychometric module reduces Adaptivity and Tone wins with PsyScore retaining above 75% — related
• DeepTutor improves overall first-person interactive tutoring quality by 10.76% over a Naive Tutor baseline on TutorBench — related