Author(s)

Shuo Sun

Corresponding Author

Shuo Sun

Abstract

With the rapid development of artificial intelligence, especially generative AI based on Transformer models, personalized learning has gained significant attention in the field of intelligent education. This paper proposes a framework for optimizing learning paths using generative AI, integrating the principles of Cognitive Load Theory (CLT) to create dynamic, adaptable learning experiences. The framework consists of four key modules: learner profiling, cognitive load analysis, Transformer-based path generation, and feedback adaptation. By dynamically adjusting learning paths based on real-time cognitive load assessments, this system enhances learning efficiency and learner engagement. The integration of CLT ensures that the cognitive load is optimally balanced, preventing overload and promoting deeper learning. Through a series of empirical experiments, the proposed system is validated, demonstrating its potential to improve learning outcomes in personalized educational environments. This work contributes to advancing the field of AI-driven education systems, offering a novel approach to adaptive learning path generation that is both cognitively informed and highly responsive to individual learner needs.

Keywords

Generative AI, Learning Path Optimization, Transformer, Cognitive Load Theory, Personalized Learning, Adaptive Learning Systems, Educational Technology, Intelligent Education, Cognitive Load Regulation, Feedback Adaptation