Classroom Group Attention Recognition and Feedback Supported by GenAI: Model Construction and Practical Pathways

With the widespread application of generative artificial intelligence (GenAI) in education, classroom teaching is undergoing a profound transformation from experience-driven to data-driven and human-machine collaboration. How to realize the precise perception and efficient feedback of learning states has become a critical issue for intelligent instructional decisionmaking. However, existing studies predominantly focus on individual attention assessment while neglecting group interaction effects, generally revealing the limitation of “valuing recognition over feedback”.Therefore, taking classroom group attention as the entry point, this paper systematically analyzed the demands for group attention monitoring and teachers’feedback, and proposed an intelligent feedback model integrating recognition, analysis, reasoning, and feedback, thereby promoting the transformation of GenAI from tool-enabled empowerment to partner-oriented support. The study aimed to provide theoretical foundations and practical pathways for constructing a perceptive, interpretable and collaborative classroom teaching feedback system, and validated its feasibility through empirical research, thereby expanding the value boundary of generative artificial intelligence in classroom instruction.