Personalized Feedback Based on Artificial Intelligence: A Model for a Humanities Master’s

The purpose of the article is to present and justify a pedagogical model of personalized feedback based on large language models (LLM) for the educational process in a humanities-oriented master’s program. The relevance of the study is determined by the objectives of digital transformation of higher education in the Russian Federation, outlined in Presidential Decree No. 474 dated July 21, 2021, “On National Development Goals until 2030”, and the need to implement flexible and adaptive educational technologies using artificial intelligence in the pedagogical process. The research methodology was based on an analysis of scientific literature and regulatory documents, as well as on the design and testing of the Master’s program “School Management and Educational Policy” at the Moscow City Pedagogical University in the 2024-2025 academic year (n = 52). The result is a comprehensive model that integrates a task-modular approach, a system for solving learning tasks, a competence assessment matrix, and automated generation of structured feedback based on LLM using prompt engineering. During the validation process, 247 solutions were obtained, and Rasch modeling showed acceptable fit of the AI-generated data to the model at the level of individual indicators; AI feedback on master’s students’ performance was no worse than feedback provided by the program’s instructors. Key conclusion: the model allows for prompt, criterion-based feedback, contributing to the personalization of learning trajectories and the development of professional competence of master students. The model can be scaled to other humanities training areas.