Comparing the performance of ChatGPT-3.5-Turbo, ChatGPT-4, and Google Bard with Iranian students in pre-internship comprehensive exams
Soolmaz Zare, Soheil Vafaeian, Mitra Amini, Keyvan Farhadi, Mohammadreza Vali, Ali Golestani · 2024
Read Comparing the performance of ChatGPT-3.5-Turbo, ChatGPT-4, and Google Bard with Iranian students in pre-internship comprehensive exams by Soolmaz Zare, Soheil Vafaeian, Mitra Amini, Keyvan Farhadi, Mohammadreza Vali, Ali Golestani (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-11574257.txt.
Description
Soolmaz Zare and co-authors test AI chatbots on medical exams in a 2024 paper. Three sets of Iranian pre-internship exams were given, in Persian and English, to two GPT models and Google Bard, and scores were set against Iranian medical students. Visual questions were left out, and each query ran in a fresh session.
GPT-4 topped every exam in both languages. Bard passed in Persian with below-average scores, while the smaller GPT model failed all sets. On English exams, model gaps over students were not firm. Readers get evidence that the strongest model can beat students on text exams, with limits still to map.
Catalog record
| CREATOR | Soolmaz Zare, Soheil Vafaeian, Mitra Amini, Keyvan Farhadi, Mohammadreza Vali, Ali Golestani |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 37 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |