Clinical NLP

Processing clinical notes and medical text.

1
Datasets
3
Results
Accuracy
Canonical metric
Canonical Benchmark

MedQA (USMLE)

Multiple-choice medical question answering dataset derived from US Medical Licensing Exam (USMLE) practice questions.

Primary metric: Accuracy
View full leaderboard

Top 10

Leading models on MedQA (USMLE).

RankModelAccuracyYearSource
1
Med-Gemini
91.12026paper
2
Med-PaLM 2
86.52026paper
3
GPT-4 (base)
86.12026paper

All datasets

1 dataset tracked for this task.

Related tasks

Other tasks in Medical.