A Survey on Retrieval-Augmented Generation in Applications of Education and Teaching
Feng Pan, Qiyun Zhou, Weitong Guo, Hongwu Yang · 2025
Retrieval-augmented generation (RAG) seamlessly combines information retrieval with generative artificial intelligence, allowing large language models (LLMs) to enhance the accuracy and relevance of responses by dynamically retrieving and integrating contextual knowledge from external databases. In education, RAG holds transformative potential, facilitating personalized learning path recommendations, automated question-answering systems, knowledge retrieval, generation of teaching resources, and adaptive evaluation frameworks. Its strengths, such as efficiency, precision, personalization, and interactivity, streamline the acquisition and utilization of educational resources, ultimately fostering improved pedagogical outcomes. Nonetheless, several challenges remain, such as integrating interdisciplinary knowledge, addressing generative hallucinations, and navigating ethical considerations in data usage. Future advancements are expected to concentrate on precision teaching support, adaptive learning ecosystems, data-driven educational governance, and specialized applications in areas like vocational training and inclusive education. RAG is positioned to reshape educational practices by addressing technical and pedagogical barriers and promoting equitable access and personalized learning experiences in an evolving digital landscape.