NLP Methods May Actually Be Better Than Professors at Estimating Question Difficulty

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zotos, Leonidas, de Jong, Ivo Pascal, Valdenegro-Toro, Matias, Sburlea, Andreea Ioana, Nissim, Malvina, van Rijn, Hedderik
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911269453299712
author Zotos, Leonidas
de Jong, Ivo Pascal
Valdenegro-Toro, Matias
Sburlea, Andreea Ioana
Nissim, Malvina
van Rijn, Hedderik
author_facet Zotos, Leonidas
de Jong, Ivo Pascal
Valdenegro-Toro, Matias
Sburlea, Andreea Ioana
Nissim, Malvina
van Rijn, Hedderik
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2508_03294
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle NLP Methods May Actually Be Better Than Professors at Estimating Question Difficulty
Zotos, Leonidas
de Jong, Ivo Pascal
Valdenegro-Toro, Matias
Sburlea, Andreea Ioana
Nissim, Malvina
van Rijn, Hedderik
Computation and Language
Artificial Intelligence
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.
title NLP Methods May Actually Be Better Than Professors at Estimating Question Difficulty
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2508.03294