Towards Just-in-Time Adaptive Feedback: Enhancing Student Learning via Knowledge-Grounded LLM

2026-05-25

AI Summary

This study addresses the lack of immediate, adaptive feedback mechanisms for addressing students’ conceptual misconceptions in current instructional settings. The authors propose a knowledge-enhanced large language model (LLM) approach that analyzes the reasoning embedded in students’ strategic explanations to automatically identify error types and generate non-intrusive, just-in-time conceptual clarification feedback. This method represents the first deployment of a real-time feedback system—integrating expert knowledge with LLM capabilities—in a large-scale university classroom involving over a thousand students. Experimental results demonstrate that student performance in the course improved by more than 80% compared to previous cohorts, and learning trajectory analyses further confirm the system’s effectiveness in facilitating conceptual change.

Abstract

Educational interventions are effective tools for enhancing student learning. While Large Language Models (LLMs) allow for generating adaptive feedback at scale, current studies lack clear methodologies for providing Just-in-Time (JiT) feedback in authentic instructional settings. In this paper, we present a framework that provides adaptive feedback by grounding LLMs with domain-specific expert knowledge. Our approach collects written reasoning logic (strategy essays) from students, analyzes potential error types based on the content of that reasoning, and delivers non-intrusive feedback designed to clarify missing or incorrect concepts. We deploy this framework in a large-scale university course (N > 1000), where it improved student performance by over 80% compared to previous semesters. Lastly, we validate the framework's pedagogical utility by analyzing the learning trajectories; we demonstrate how iterative conversations with LLM facilitate shifting one's misconception to correct understanding.

Problem

• Just-in-Time feedback
• adaptive feedback
• student learning
• Large Language Models
• educational intervention

Innovation

• Just-in-Time feedback
• Knowledge-Grounded LLM
• Adaptive Educational Intervention
• Student Reasoning Analysis
• Misconception Remediation