Automatic questions generation based on keywords using language models

Authors

  • Tomasz Gniazdowski JT Weston sp. z o.o. Warszawa, Poland
  • Marek Bazan Wrocaw University of Science and Technology, Faculty of Information and Communication Technology, Department of Computer Engineering, Poland
  • Maciej E. Marchwiany Interdisciplinary Centre for Mathematical and Computational Modelling (ICM), University of Warsaw, Poland

DOI:

https://doi.org/10.24425/ijet.2025.153615

Abstract

In this paper, we presented a novel method for question generation, one of the most impactful NLP tasks in contacts of user interfaces, chatbots and intelligent assistants with a user. Our method outperforms commonly used methods in terms of quality and speed of question generation. Additionally, we benchmarked the most used methods of question generation that are based on the usage of Large Language Models in a fewshot approach as well as finetuned to that task. Our work is done for the Polish language, it has one of the most challenging and complex grammars, which makes the task even more difficult.

Published

2025-07-11

How to Cite

Gniazdowski, Tomasz, et al. “Automatic Questions Generation Based on Keywords Using Language Models”. International Journal of Electronics and Telecommunications, vol. 71, no. 3, July 2025, doi:10.24425/ijet.2025.153615.

Issue

Section

Artykuły