Evaluating Bard Gemini Pro and GPT-4 Vision Against Student Performance in Medical Visual Question Answering: Comparative Case Study
Jonas Roos, Ron Martin, Robert Kaczmarczyk · 2024
Read Evaluating Bard Gemini Pro and GPT-4 Vision Against Student Performance in Medical Visual Question Answering: Comparative Case Study by Jonas Roos, Ron Martin, Robert Kaczmarczyk (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-11683658.txt.
Description
Jonas Roos, Ron Martin, and Robert Kaczmarczyk compare AI chatbots with students on medical image questions in a 2024 study. The test held over one thousand image based exam questions in English and German. Each model had to read the image and pick the most likely diagnosis. Student scores came from the same learning platform.
One model answered about fifty-seven percent correctly, ahead of its rival at about forty-five percent. It also skipped far more questions than the rival did. On answered items alone, it beat the average student pass rate. A student majority vote still won by a wide margin, near ninety-five percent. Both models scored better in German than in English.
Catalog record
| CREATOR | Jonas Roos, Ron Martin, Robert Kaczmarczyk |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |