Retrieval-Augmented Tutoring for Algorithm Tracing and Problem-Solving in AI Education

2026-05-13

AI Summary

Students commonly struggle with understanding algorithmic execution traces, debugging reasoning errors, and transferring knowledge to new contexts. To address these challenges, this work proposes KITE, an intelligent tutoring system that innovatively integrates retrieval-augmented generation (RAG) with an intention-aware Socratic dialogue strategy. KITE delivers precise and pedagogically appropriate interactive feedback by combining course-content retrieval, scaffolded instructional prompts, and guided questioning. The system employs a multimodal RAG pipeline and leverages large language models to simulate student interactions, alongside a multidimensional evaluation framework. Experimental results demonstrate that KITE-generated feedback excels in curricular alignment and instructional effectiveness, significantly improving student model accuracy on algorithm tracing and procedural problem-solving tasks.

Abstract

Students learning algorithms often need support as they interpret traces, debug reasoning errors, and apply procedures across unfamiliar problem instances. In this paper, we present KITE (Knowledge-Informed Tutoring Engine), a Retrieval-Augmented Generation (RAG)-based intelligent tutoring system designed to serve as a classroom teaching assistant for algorithmic reasoning and problem-solving tasks. KITE uses an intent-aware Socratic response strategy to tailor support to different student needs, responding with targeted hints, guiding questions, and progressive scaffolding intended to strengthen students' algorithmic problem-solving ability. To keep responses aligned with course content, KITE uses a multimodal RAG pipeline that retrieves relevant information from course materials. We evaluate KITE using three forms of assessment: RAGAs-based metrics for response grounding and quality, expert evaluation of pedagogical quality, and a simulated student pipeline in which a weaker language model interacts with KITE across two-turn dialogues and produces revised answers after receiving feedback. Results indicate that KITE produces contextually grounded and pedagogically appropriate responses. Further, using simulated students, KITE's feedback helped the student models produce more accurate follow-up responses on procedural and tracing questions, suggesting that its scaffolding can support algorithmic problem-solving. This work contributes a tutoring architecture and an evaluation approach for assessing retrieval-grounded explanations and scaffolded problem-solving feedback.

Problem

• algorithm tracing
• problem-solving
• intelligent tutoring
• reasoning errors
• algorithmic reasoning

Innovation

• Retrieval-Augmented Generation
• Intelligent Tutoring System
• Socratic Dialogue
• Algorithmic Problem-Solving
• Multimodal RAG