Constructing a Generative AI Assistant for the Reform of University Experimental Teaching: A Case Study of the Advanced Language Programming (C Language) Course
Keywords:
Generative Artificial Intelligence (GenAI), Advanced Language Programming, Large Language Model (LLM)Abstract
With the rapid advancement of information technology, Advanced Language Programming (C Language) has become a core curriculum in computer-related disciplines in higher education institutions. However, the course's intrinsic characteristics—such as its strong practicality, rapid technological updates, and significant variance in student foundational knowledge—pose considerable challenges to traditional experimental teaching. To address these challenges and enhance instructional quality and personalized learning experience, this paper proposes and implements a Generative AI Teaching Assistant System (GenAI-TA) specifically designed for the "Advanced Language Programming (C Language)" experimental course. The system is built upon an advanced Large Language Model (LLM) and integrates Retrieval-Augmented Generation (RAG) technology. It is fine-tuned by incorporating a course-specific knowledge base (including the syllabus, lab manuals, code examples, and a collection of common errors) to provide precise, real-time, and personalized tutoring. This paper elaborates on the GenAI-TA's system architecture, key technical implementation details, and its integration scheme with the Geely University OJ Platform development environment. To evaluate its pedagogical effectiveness, a one-semester quasi-experimental study was conducted. The results indicate that, compared to the control group under the traditional teaching model, students in the experimental group using GenAI-TA achieved significant improvements in programming skills, project completion quality, and problem-solving abilities (p < 0.05). Furthermore, the System Usability Scale (SUS) score of 85.7 suggests a high level of student acceptance and satisfaction with the system. This research validates the immense potential of Generative AI teaching assistants in reforming the experimental instruction of practical courses and provides empirical evidence and a feasible technical pathway for the deeper application of AI in the higher education sector.
References
Krusche, S., & Alperowitz, L. (2018). Introduction of continuous delivery in multi-customer project courses. In Proceedings of the 40th International Conference on Software Engineering: Software Engineering Education and Training (pp. 31–40). IEEE. https://doi.org/10.1145/3183377.3183378
Villegas-Ch, W., Román-Cañizares, M., & Palacios-Pacheco, X. (2020). Improvement of an education online model with the integration of machine learning and data analysis in an institution of higher education. Sensors, 20(5), Article 1396. https://doi.org/10.3390/s20051396
Chiu, T. K. F. (2023). The impact of generative AI (GenAI) on practices of higher education: Challenges and opportunities. Smart Learning Environments, 10, Article 39. https://doi.org/10.1186/s40561-023-00258-w
Kilde-Westberg, S., & Bøe, M. V. (2025). Generative AI as a lab partner: A case study in a university physics course. Computers and Education: Artificial Intelligence, 8, Article 100344. https://doi.org/10.1016/j.caeai.2024.100344
Lau, S., & Guo, P. J. (2023). Banter: A pedagogically-appropriate AI tutor for computer science education. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems. ACM. https://doi.org/10.1145/3544548.3581564
Essel, H. B., Vlachopoulos, D., Tachie-Menson, A., Johnson, E. E., & Baah, P. K. (2024). The impact of generative AI (ChatGPT) on university students’ engagement and learning gains: A systematic review. Higher Education Quarterly. Advance online publication. https://doi.org/10.1111/hequ.12504
Brooke, J. (1996). SUS: A "quick and dirty" usability scale. In P. W. Jordan, B. Thomas, B. A. Weerdmeester, & I. L. McClelland (Eds.), Usability evaluation in industry (pp. 189–194). Taylor & Francis.
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., ... & Kiela, D. (2020). Retrieval-augmented generation for knowledge-intensive NLP tasks. In Advances in Neural Information Processing Systems 33 (pp. 9459–9474). Curran Associates, Inc.
Hu, E. J., Shen, Y., Wallis, P., Allen-Zhu, Z., Li, Y., Wang, S., Wang, L., & Chen, W. (2021). LoRA: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685.
Barke, S., James, M. B., & Polikarpova, N. (2023). Grounded copilot: How programmers interact with code-generating models. Proceedings of the ACM on Programming Languages, 7(OOPSLA2), 85–111. https://doi.org/10.1145/3618307
Holmes, W., Persson, J., Chounta, I. R., Wasson, B., & Dimitrova, V. (2022). Ethics of artificial intelligence in education: Towards a community-wide framework. International Journal of Artificial Intelligence in Education, 32(3), 504–526. https://doi.org/10.1007/s40593-022-00292-9
Farrokhnia, M., Konijn, E. A., Akyüz, N., & Beer, N. (2024). A favourable buddy or a fearful beast? A systematic review of ChatGPT's roles, benefits, and challenges in education. Educational Technology Research and Development, 72(1), 1–43. https://doi.org/10.1007/s11423-023-10298-w
