Mined misconception labels match personas' assigned misconceptions at F1 ≈ 0.56, a score that does not test whether generated questions elicit the named misconception¶

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? The grader's recovery of persona misconception labels from learner answers reaches F1 ≈ 0.56, computed post hoc by token-overlap matching over all 132 rounds. → Kailai He 2026

Evidence¶

Kailai He 2026¶

Kailai He, Zhihao Wu, Linhai Zhang, Runcong Zhao, Yulan He, Jiazheng Li. (2026). CoLearn: An Agentic Tutor that Learns its Learner in a Human–AI Co-Learning Loop. https://arxiv.org/abs/2609.21154 doi:10.48550/arxiv.2609.21154

q2 · i? · design · r2

Persona-simulation evaluation of the grader's misconception mining; the article reports "Mined misconceptions match the per- sonas’ assigned ones at F1≈0.56", computed post hoc by token-overlap matching over all 132 rounds.

"Mined misconceptions match the per- sonas’ assigned ones at F1≈0.56"

Discussion¶

Related Claims¶

• Per-topic performance varied: NP-completeness scored highest ROUGE-1 F1 (0.1285), sorting algorithms highest pedagogical quality (0.8250), and recurrence relations lowest pedagogical quality (0.6629) — related
• CogEvolution performs comparably to the KT-based PEERS model on mastery prediction (AUC 0.80 vs 0.82) while achieving higher mistake precision (76.8%) on CogMath-948 — related
• Frontier LLMs achieve the best balance between detecting misconceptions in correct-answer cases and not over-flagging correct students, while fine-tuned local models achieve under 58% TM recall — related