A benchmark study reports GPT-5 outperforming GPT-4o and pre-licensed human experts on most medical QA tasks, but not consistently on VQA-RAD.
Assessing the accuracy and reliability of ai-generated medical responses: an evaluation of the chat-gpt model
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
REJECT 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Capabilities of GPT-5 on Multimodal Medical Reasoning
A benchmark study reports GPT-5 outperforming GPT-4o and pre-licensed human experts on most medical QA tasks, but not consistently on VQA-RAD.