CLLMRec: LLM-powered Cognitive-Aware Concept Recommendation via Semantic Alignment and Prerequisite Knowledge Distillation

TL;DR

CLLMRec addresses MOOC personalized concept recommendation by removing reliance on external knowledge graphs and instead leveraging LLMs through Semantic Alignment and Prerequisite Knowledge Distillation. A teacher LLM extracts latent prerequisite relations to generate soft labels, which train a lightweight Student Ranker, while a Fine-Ranker integrates real-time cognitive states from DKT for final calibration. The framework uses frozen LLM encoders with prompts that yield unified semantic representations and distills structured guidance into an end-to-end objective combining distillation, preference learning, and re-ranking. Experiments on ASSIST09 and ASSIST12 demonstrate substantial gains over strong baselines, validating cognitive-aware, graph-free recommendations with practical impact for scalable MOOC personalization.

Abstract

The growth of Massive Open Online Courses (MOOCs) presents significant challenges for personalized learning, where concept recommendation is crucial. Existing approaches typically rely on heterogeneous information networks or knowledge graphs to capture conceptual relationships, combined with knowledge tracing models to assess learners' cognitive states. However, these methods face significant limitations due to their dependence on high-quality structured knowledge graphs, which are often scarce in real-world educational scenarios. To address this fundamental challenge, this paper proposes CLLMRec, a novel framework that leverages Large Language Models through two synergistic technical pillars: Semantic Alignment and Prerequisite Knowledge Distillation. The Semantic Alignment component constructs a unified representation space by encoding unstructured textual descriptions of learners and concepts. The Prerequisite Knowledge Distillation paradigm employs a teacher-student architecture, where a large teacher LLM (implemented as the Prior Knowledge Aware Component) extracts conceptual prerequisite relationships from its internalized world knowledge and distills them into soft labels to train an efficient student ranker. Building upon these foundations, our framework incorporates a fine-ranking mechanism that explicitly models learners' real-time cognitive states through deep knowledge tracing, ensuring recommendations are both structurally sound and cognitively appropriate. Extensive experiments on two real-world MOOC datasets demonstrate that CLLMRec significantly outperforms existing baseline methods across multiple evaluation metrics, validating its effectiveness in generating truly cognitive-aware and personalized concept recommendations without relying on explicit structural priors.

Figures (1)

• Figure 1: Overall Architecture and Data Flow of CLLMRec. The diagram illustrates the end-to-end pipeline: user and concept data are encoded by the LLM, processed by the ranker which fuses prerequisites and preferences, and finally calibrated with cognitive states to produce recommendations.