AUTOMATIC MULTIPLE-CHOICE QUESTION GENERATION USING ARTIFICIAL INTELLIGENCE

Abstract

This paper presents a local pipeline for generating multiple-choice questions for IGCSE and A-Level computer science. From 3,385 source questions, duplicate removal and answer-position balancing produced 5,702 training and validation examples. Qwen2.5-7B-Instruct was adapted using QLoRA on an NVIDIA RTX 3060 with 12 GB of memory, and generated candidates underwent automatic structural validation. An author audit judged 15 of 18 candidates usable without substantive correction. A subsequent external expert review rated 51 of 57 items (89.5%) usable as generated and six usable after minor revision; no factual errors were recorded. Answer leakage and repeated stems remained important defects. The study documents the data-preparation procedure, training configuration, and evaluation limitations. The findings support a teacher-in-the-loop proof of concept, while small samples, limited rater independence, missing controlled baselines, and unresolved source-data permissions constrain broader claims.

Keywords: automatic question generation, multiple-choice questions, large language models, QLoRA, computer science education.

Список литературы

1. Kurdi G., Leo J., Parsia B., Sattler U., Al-Emari S. A systematic review of automatic question generation for educational purposes // Int. J. Artif. Intell. Educ. — 2020. — Vol. 30. — P. 121–204. DOI: 10.1007/s40593-019-00186-y.
2. Ch D.R., Saha S.K. Automatic multiple choice question generation from text: A survey // IEEE Trans. Learn. Technol. — 2020. — Vol. 13, no. 1. — P. 14–25. DOI: 10.1109/TLT.2018.2889100.
3. Heilman M., Smith N.A. Good question! Statistical ranking for question generation // Human Language Technologies: The 2010 Annual Conference of the North American Chapter of the Association for Computational Linguistics, Association for Computational Linguistics. — 2010. — P. 609–617. URL: https://aclanthology.org/N10-1086/.
4. Du X., Shao J., Cardie C. Learning to ask: Neural question generation for reading comprehension // Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), Association for Computational Linguistics. — 2017. — P. 1342–1352. DOI: 10.18653/v1/P17-1123.
5. Kasneci E., Seßler K., Küchemann S., et al. ChatGPT for good? On opportunities and challenges of large language models for education // Learn. Individ. Differ. — 2023. — Vol. 103. — Article 102274. DOI: 10.1016/j.lindif.2023.102274.
6. Doughty J., Wan Z., Bompelli A., et al. A comparative study of AI-generated (GPT-4) and human-crafted MCQs in programming education // Proceedings of the 26th Australasian Computing Education Conference, ACM. — 2024. — P. 114–123. DOI: 10.1145/3636243.3636256.
7. Vaswani A., Shazeer N., Parmar N., Uszkoreit J., Jones L., Gomez A.N., Kaiser Ł., Polosukhin I. Attention is all you need // Advances in Neural Information Processing Systems. — 2017. — Vol. 30. — P. 5998–6008. DOI: 10.48550/arXiv.1706.03762.
8. Qwen Team. Qwen2.5 technical report [Electronic resource]. — arXiv:2412.15115v1, 2024. URL: https://arxiv.org/abs/2412.15115v1.
9. Hu E.J., Shen Y., Wallis P., Allen-Zhu Z., Li Y., Wang S., Wang L., Chen W. LoRA: Low-rank adaptation of large language models // International Conference on Learning Representations, 2022. URL: https://openreview.net/forum?id=nZeVKeeFYf9.
10. Dettmers T., Pagnoni A., Holtzman A., Zettlemoyer L. QLoRA: Efficient finetuning of quantized LLMs // Advances in Neural Information Processing Systems. — 2023. — Vol. 36. — P. 10088–10115. DOI: 10.52202/075280-0441.
11. Wolf T., Debut L., Sanh V., et al. Transformers: State-of-the-art natural language processing // Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing: System Demonstrations, Association for Computational Linguistics. — 2020. — P. 38–45. DOI: 10.18653/v1/2020.emnlp-demos.6.
12. Hugging Face. PEFT documentation [Electronic resource]. — 2026. URL: https://huggingface.co/docs/peft/ (accessed: 15.07.2026).
13. Hugging Face. Transformers documentation [Electronic resource]. — 2026. URL: https://huggingface.co/docs/transformers/ (accessed: 15.07.2026).
14. Dettmers T., Lewis M., Belkada Y., Zettlemoyer L. LLM.int8(): 8-bit matrix multiplication for transformers at scale // Advances in Neural Information Processing Systems. — 2022. — Vol. 35. — P. 30318–30332. DOI: 10.52202/068431-2198.
15. Raffel C., Shazeer N., Roberts A., Lee K., Narang S., Matena M., Zhou Y., Li W., Liu P.J. Exploring the limits of transfer learning with a unified text-to-text transformer // J. Mach. Learn. Res. — 2020. — Vol. 21, no. 140. — P. 1–67. URL: https://jmlr.org/papers/v21/20-074.html.
16. Chung H.W., Hou L., Longpre S., et al. Scaling instruction-finetuned language models // J. Mach. Learn. Res. — 2024. — Vol. 25, no. 70. — P. 1–53. URL: https://jmlr.org/papers/v25/23-0870.html.

Как цитировать

Amirlan Seitkadyrov, Akezhan Omashev AUTOMATIC MULTIPLE-CHOICE QUESTION GENERATION USING ARTIFICIAL INTELLIGENCE // Central Asian Scientific Journal. – 2026. – No. 4(32), Vol. 1. – Pp. 287–295.