Question-answering and dialogue histories are complementary views of the same student, with signals transferring across metric categories¶

Claim · All claims
Evidence · 1 study · causal 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? Question-answering records help knowledge metrics but also dialogue-act prediction, and dialogue histories add fine-grained knowledge information beyond linguistic style. → Zhangqi Duan 2026

Evidence¶

Zhangqi Duan 2026¶

Zhangqi Duan, Shuyan Huang, Alexander Scarlatos, Jaewook Lee, Simon Woodhead, Andrew Lan. (2026). Who Am I? History-Aware Profiles for Student Simulation in Tutoring Dialogues. arXiv preprint. https://arxiv.org/abs/2605.30051 doi:10.48550/arxiv.2605.30051

q2 · i? · causal · r2

Analysis of main experiment results (Section 5.1). The article explains that students' knowledge levels shape dialogue behavior, e.g. high-knowledge students seek help less often, and prior tutor scaffolding reveals how students respond to hints.

"Most importantly, question-answering and dia- logue histories provide complementary signals that transfer across metric categories, rather than only improving the most targeted metrics."

Discussion¶

Related Claims¶

• Knowledge and behavior profile components provide complementary signals: removing either hurts metrics tied to the other — related
• GRPO training shifts profiles from broad summaries toward diagnostic, behaviorally grounded representations that better support simulation — related