In a pilot across 39 classrooms, teacher-authored prompts produced 71% on-track student-AI conversations but 38% under-reached the targeted cognitive demand level, with explicit finish lines cutting the gap by 0.22 levels and no-direct-answer guardrails cutting final-answer rates by 8.5 points.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Teacher-Authored Prompts for Configuring Student-AI Dialogue: K-12 Classroom Implementation
In a pilot across 39 classrooms, teacher-authored prompts produced 71% on-track student-AI conversations but 38% under-reached the targeted cognitive demand level, with explicit finish lines cutting the gap by 0.22 levels and no-direct-answer guardrails cutting final-answer rates by 8.5 points.