The American surgeon
Advances in Large Language Model Performance: A Comparative Study of ChatGPT-4 and ChatGPT-5 on ABSITE Questions.
Ahmet Necati Sanli, Bilal Turan, Deniz Esin Tekcan Sanli
Published: 202510.1177/00031348251390958
Abstract
ObjectiveThis study aimed to directly compare the performance of 2 successive versions of ChatGPT (ChatGPT-4o and ChatGPT-5) in answering questions from the American Board of Surgery In-Training Examination (ABSITE) Quiz.MethodsA total of 170 multipl…
Preview only. Read the full abstract at the source