MedQA-USMLE
A medical question-answering benchmark constructed from United States Medical Licensing Examination (USMLE) questions. It is used to evaluate models' complex relational reasoning and knowledge application in the medical domain.
Timeline (last 90 days)
VisKG-LM achieves improvements of 1.2, 0.8, and 4.3 points over the GreaseLM baseline on the three benchmarks CommonsenseQA, OpenBookQA, and MedQA-USMLE, respectively.
MedQA-USMLE is drawn from the United States Medical Licensing Examination and requires multi-step reasoning over complex relationships between medical concepts.
The improvement of VisKG-LM on MedQA-USMLE is particularly significant, partly because medical knowledge graphs have high relation density and long reasoning chains, making them well-suited for visual 2D layouts that preserve multi-path crossing structures.
All Facts (3)
VisKG-LM achieves improvements of 1.2, 0.8, and 4.3 points over the GreaseLM baseline on the three benchmarks CommonsenseQA, OpenBookQA, and MedQA-USMLE, respectively.
50%UnverifiedMedQA-USMLE is drawn from the United States Medical Licensing Examination and requires multi-step reasoning over complex relationships between medical concepts.
50%UnverifiedThe improvement of VisKG-LM on MedQA-USMLE is particularly significant, partly because medical knowledge graphs have high relation density and long reasoning chains, making them well-suited for visual 2D layouts that preserve multi-path crossing structures.
50%