Research on the Reform of Teaching Compiler Principles Empowered by GraphRAG

Abstract

With the rapid advancement of artificial intelligence technology, traditional teaching models are undergoing profound transformation. Compiler Theory courses are widely regarded as one of the most challenging in computer science due to their abstract theoretical concepts, tightly interconnected knowledge systems, and significant gap between theory and practice. This creates substantial challenges for students’ learning process and makes it difficult for instructors to achieve expected teaching objectives. To address these challenges, this paper applies Graph-Augmented Retrieval Generation (GraphRAG) technology to teaching compiler theory. By integrating knowledge graphs with large language models, it visualizes knowledge, enhances AI’s personalized semantic question-answering capabilities, and reduces students’ knowledge barriers in understanding abstract concepts, thereby achieving intelligent teaching assistance. Through teaching practice, we validated student learning outcomes. Results indicate this method enhances students’ comprehension of abstract concepts and improves their self-directed learning abilities to a certain extent, demonstrating practical value.