Research on Hybrid Teaching Design and Interaction Mechanism Optimization in Higher Education Based on the DeepSeek Large Language Model

Abstract

This study explores the Applied Value and optimization effect of the deepseek Big language model in a mixed learning scenario in higher education institutions and creates the structure of a three-way “teacher-student-model” interaction. Empirical analysis confirmed the positive effects of the deepseek model in increasing student motivation for learning, cognitive processing of information, and reducing cognitive overload. Evidence shows that real-time intellectual interaction and personalized feedback significantly increased academic efficiency, autonomy, and cognitive control, as well as shifting the role of traditional teachers from simply transferring knowledge to leading courses, bringing new insights and directions of change to teaching methods. In addition, the degree to which knowledge-seeking people adopt new technologies is closely related to their overall use and learning efficiency. At the same time, the relationship to technology adoption also acts as an intermediate variable to determine the impact of learning outcomes on cognitive load. Relevant research results provide a practical solution for the in-depth cooperation model of an intelligent education system; and this requires the next step - continuous assessment of the task of continuous efficiency and strengthening the training of teachers in the training process to adapt to the implementation.