Automated Knowledge Gap Detection from Student-AI Chat Logs
Published:Dec 26, 2025 23:04
•1 min read
•ArXiv
Analysis
This paper proposes a novel approach to identify student knowledge gaps in large lectures by analyzing student interactions with AI assistants. The use of student-AI dialogues as a data source is innovative and addresses the limitations of traditional classroom response systems. The framework, QueryQuilt, offers a promising solution for instructors to gain insights into class-wide understanding and tailor their teaching accordingly. The initial results are encouraging, suggesting the potential for significant impact on teaching effectiveness.
Key Takeaways
Reference
“QueryQuilt achieves 100% accuracy in identifying knowledge gaps among simulated students and 95% completeness when tested on real student-AI dialogue data.”