Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

MedQuAD

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringMedQUAD
PRR58.3
66
Selective GenerationMedQUAD
ROC-AUC0.928
66
Selective GenerationMedQUAD
PRR (ROUGE-L)46.6
14
Selective GenerationMedQUAD Out-of-domain
PRR (ROUGE-L)30.4
8
Medical Question AnsweringMedQuAD-style Complete Benchmark
MedQuAD Score91.1
5
Medical Question AnsweringMedQuAD (test)
ROUGE-153
4
Medical ReasoningMedQuAD complete benchmark
Failure Rate0
1
Showing 7 of 7 rows