LARGE LANGUAGE MODELS AS A TOOL IN PHYSICS TRAINING OF FUTURE ENGINEERS

Abstract

The article is devoted to the study of the possibilities of large language models in teaching physics to engineering students. The relevance is conditioned by the necessity of personalisation of physics teaching and selection of technologies adequate to the tasks of knowledge-intensive production, digital economy and cognitive abilities of modern students. The aim of the study is to analyse the functionality and identify the advantages of large language models in personalisation and efficiency of physics teaching. The research methodology is based on the concept of lifelong learning, open learning model of physics, blended learning method, and empirical method of studying the didactic possibilities of large language models. The author proposed the methods of personalisation of the educational process, ways of developing a new type of tasks related to the application of large language models, recommendations on formulating criteria for evaluating the results of learning activities are. It was demonstrated that GPT-4o and Claude 3.5 Sonnet in the course of contextual dialogue can explain solutions to most physics problems offered to engineering students, but under the condition of critical analysis of all stages of the solution. The practical significance of the work lies in the development of methods of using large language models to automate the procedure of test creation and their embedding in Moodle, visualization of physical processes, organization of interactive classes in the format of blended learning. New pedagogical methods such as: competitions using large language models, contextual dialogue with neural networks, using their computational capabilities in carrying out similar calculations are of applied value. The conclusion emphasises the need to adapt educational programmes to the capabilities of large language models and to conduct further research into their didactic potential.