LLM-driven Effective Knowledge Tracing by Integrating Dual-channel Difficulty

TL;DR

This paper introduces a novel LLM-based framework for knowledge tracing that effectively addresses cold-start issues, improves personalization accuracy, and enhances interpretability through dual-channel difficulty assessment and mastery modeling.

Contribution

The paper proposes the DDKT framework integrating LLMs, RAG, and difficulty-aware algorithms to improve knowledge tracing accuracy and interpretability, especially for new questions.

Findings

Outperforms nine baseline models with 2-10% AUC improvement.

Effectively alleviates cold-start problems in knowledge tracing.

Enhances model interpretability through difficulty perception mechanisms.

Abstract

Knowledge Tracing (KT) is a fundamental technology in intelligent tutoring systems used to simulate changes in students' knowledge state during learning, track personalized knowledge mastery, and predict performance. However, current KT models face three major challenges: (1) When encountering new questions, models face cold-start problems due to sparse interaction records, making precise modeling difficult; (2) Traditional models only use historical interaction records for student personalization modeling, unable to accurately track individual mastery levels, resulting in unclear personalized modeling; (3) The decision-making process is opaque to educators, making it challenging for them to understand model judgments. To address these challenges, we propose a novel Dual-channel Difficulty-aware Knowledge Tracing (DDKT) framework that utilizes Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) for subjective difficulty assessment, while integrating difficulty bias-aware algorithms and student mastery algorithms for precise difficulty measurement. Our framework introduces three key innovations: (1) Difficulty Balance Perception Sequence (DBPS) - students' subjective perceptions combined with objective difficulty, measuring gaps between LLM-assessed difficulty, mathematical-statistical difficulty, and students' subjective perceived difficulty through attention mechanisms; (2) Difficulty Mastery Ratio (DMR) - precise modeling of student mastery levels through different difficulty zones; (3) Knowledge State Update Mechanism - implementing personalized knowledge acquisition through gated networks and updating student knowledge state. Experimental results on two real datasets show our method consistently outperforms nine baseline models, improving AUC metrics by 2% to 10% while effectively addressing cold-start problems and enhancing model interpretability.