NLP Methods May Actually Be Better Than Professors at Estimating Question Difficulty
Abstract
Estimating the difficulty of exam questions is essential for developing good exams, but professors are not always good at this task. We compare various Large Language Model-based methods with three professors in their ability to estimate what percentage of students will give correct answers on True/False exam questions in the areas of Neural Networks and Machine Learning. Our results show that the professors have limited ability to distinguish between easy and difficult questions and that they are outperformed by directly asking Gemini 2.5 to solve this task. Yet, we obtained even better results using uncertainties of the LLMs solving the questions in a supervised learning setting, using only 42 training samples. We conclude that supervised learning using LLM uncertainty can help professors better estimate the difficulty of exam questions, improving the quality of assessment.
Cite
@article{arxiv.2508.03294,
title = {NLP Methods May Actually Be Better Than Professors at Estimating Question Difficulty},
author = {Leonidas Zotos and Ivo Pascal de Jong and Matias Valdenegro-Toro and Andreea Ioana Sburlea and Malvina Nissim and Hedderik van Rijn},
journal= {arXiv preprint arXiv:2508.03294},
year = {2025}
}
Comments
10 pages, 2 figures, presented at ECAI 2025 at the 2nd International Workshop on AI in Society, Education and Educational Research (AISEER)