Tag
A study analyzing 20,462 student turns in a guided LLM chemistry tutor finds that pedagogical guardrails change what tutors withhold but not how they adapt, showing questioning helps early in impasses while direct error feedback becomes more effective as impasses persist.
This study examines how coded dialogues from GenAI virtual patients can provide teacher-interpretable process evidence of clinical reasoning in medical education, using learning analytics to analyze dialogue logs from medical students.
EduRiskX is a neuro-symbolic framework that integrates temporal Transformers with F-Logic reasoning for early prediction of academic risks in online education, demonstrating improved accuracy and interpretability compared to state-of-the-art models.
The paper shows that study-strategy clusters from EdNet logs predict learner engagement but not mastery, highlighting that behavior-only profiling is insufficient for assessing knowledge gains.
This IEEE paper presents an AI-powered knowledge graph platform that connects SQL errors to conceptual gaps in database courses, automatically extracting course concepts and linking them to student submissions to provide personalized diagnostic feedback.
The paper presents a pipeline that maps student questions from a conversational AI teaching assistant to curriculum topics using a few-shot text classifier and a GPT-4-extracted prerequisite knowledge graph, achieving 80% accuracy on 1,340 question events and correlating with self-reported difficulty.
This paper proposes an interpretable decision layer for AI-augmented classrooms that combines teacher and student feedback to rank course topics needing attention without using grades. The approach surfaces isolated learners and aligns with instructor concerns in a preliminary study.