Using small language models as tools for teaching in the ELSEI master's program

Authors

  • Aammou Souhaib Abdelmalek Essaadi University
  • Tagdimi Zakaria Abdelmalek Essaadi University
  • Touis Tarik Abdelmalek Essaadi University

Keywords:

Small language models, ELSEI, Question Answering system, Personalized Tutoring

Abstract

The Question Answering Introduction to Python (QAIP) system aims to enhance the learning experience in introductory Python courses by providing accurate and efficient answers to Python-related queries. The rise of Large Language Models (LLMs) has significantly impacted education, particularly within the framework of Education 4.0, which seeks to prepare students for a technologically advanced world. LLMs such as OpenAI’s ChatGPT and GitHub’s Copilot have revolutionized content creation, personalized tutoring, and student engagement, aligning with the goals of Education 4.0. However, the challenge of developing appropriate programming exercises and maintaining active learning in introductory programming courses persists, especially given the rapid online sharing of solutions. In this context, Small Language Models (SLMs) offer a lightweight, efficient alternative for educational integration. This article explores the integration of SLMs into the QAIP system within the E-learning and Intelligent Educational Systems (ELSEI) program, aiming to empower students with the skills to develop innovative educational tools. By narrowing the existing AI content gap, this work aspires to contribute to the broader discourse on AI accessibility and diversity. The development process involves thorough data collection, strategic model training, and careful deployment to ensure that the AI-driven system effectively meets student needs and enhances learning outcomes. Through this interdisciplinary effort, we aim to foster a culture of innovation and contribute meaningfully to the evolution of AI in education.

Author Biographies

Aammou Souhaib, Abdelmalek Essaadi University

Abdelmalek Essaadi University

Tagdimi Zakaria, Abdelmalek Essaadi University

Abdelmalek Essaadi University

Touis Tarik, Abdelmalek Essaadi University

Abdelmalek Essaadi University

References

Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., Agarwal, S., Herbert-Voss, A., Krueger, G., Henighan, T., Child, R., Ramesh, A., Ziegler, D. M., Wu, J., Winter, C., … Amodei, D. (2020). Language Models are Few-Shot Learners (arXiv:2005.14165). arXiv. http://arxiv.org/abs/2005.14165

Denny, P., Kumar, V., & Giacaman, N. (2022). Conversing with Copilot: Exploring Prompt Engineering for Solving CS1 Problems Using Natural Language (arXiv:2210.15157). arXiv. https://doi.org/10.48550/arXiv.2210.15157

Englhardt, Z., Li, R., Nissanka, D., Zhang, Z., Narayanswamy, G., Breda, J., Liu, X., Patel, S., & Iyer, V. (2024). Exploring and Characterizing Large Language Models for Embedded System Development and Debugging. Extended Abstracts of the 2024 CHI Conference on Human Factors in Computing Systems, 1–9. https://doi.org/10.1145/3613905.3650764

Gattupalli, S., Lee, W., Allessio, D., Crabtree, D., Arroyo, I., & Woolf, B. (n.d.). Exploring Pre-Service Teachers’ Perceptions of Large Language Models-Generated Hints in Online Mathematics Learning.

Grangier, D., Katharopoulos, A., Ablin, P., & Hannun, A. (2024). Specialized Language Models with Cheap Inference from Limited Domain Data (arXiv:2402.01093; Version 1). arXiv. https://doi.org/10.48550/arXiv.2402.01093

OpenAI, Achiam, J., Adler, S., Agarwal, S., Ahmad, L., Akkaya, I., Aleman, F. L., Almeida, D., Altenschmidt, J., Altman, S., Anadkat, S., Avila, R., Babuschkin, I., Balaji, S., Balcom, V., Baltescu, P., Bao, H., Bavarian, M., Belgum, J., … Zoph, B. (2024). GPT-4 Technical Report (arXiv:2303.08774). arXiv. http://arxiv.org/abs/2303.08774

Sarsa, S., Denny, P., Hellas, A., & Leinonen, J. (2022). Automatic Generation of Programming Exercises and Code Explanations using Large Language Models. Proceedings of the 2022 ACM Conference on International Computing Education Research - Volume 1, 27–43. https://doi.org/10.1145/3501385.3543957

Wang, Z., Ng, P., Ma, X., Nallapati, R., & Xiang, B. (2019). Multi-passage BERT: A Globally Normalized BERT Model for Open-domain Question Answering. In K. Inui, J. Jiang, V. Ng, & X. Wan (Eds.), Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) (pp. 5878–5882). Association for Computational Linguistics. https://doi.org/10.18653/v1/D19-1599

Published

2024-12-27

Issue

Section

For Mobilizing Communication Science for the Planet