Comparative analysis of the performance of large language models in answering endodontics questions on the dental specialty exam

View Full Paper

Ask AI

Bookmark

Share

Key Points

This study aims to compare the accuracy of different large language models in answering endodontics questions on the dental specialty exam.
Reviewed endodontics questions from Turkish dental specialty exams (2012-2021), totaling 127 questions.
Included text-based (122) and figure-based (5) items, such as periapical radiographs and clinical photographs.
Responses were scored using the official answer key and analyzed with Fisher’s exact test.
Model accuracy varied by question type, with text-based items yielding higher correct responses than figure-based.
Statistical analyses revealed significant differences in performance according to the item format type (p<0.05).
No significant variation in accuracy was noted across different exam years.

We only have this paper's title and publication details.

Compares accuracy of language models on endodontics questions, suggesting differences by item format.

No takes yet. Share an insight, caveat, or question.

Özden et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark

Share