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

ICD-Bench

Benchmarks

Task NameDataset NameSOTA ResultTrend
Question AnsweringICD-Bench filtered endocrinology
Trivial Score66.7
3
Medical ReasoningICD-Bench Difficulty Level 5 1.0
Accuracy59.05
2
Medical ReasoningICD-Bench Difficulty Level 4 1.0
Accuracy71.5
2
Medical ReasoningICD-Bench Difficulty Level 3 1.0
Accuracy80.33
2
Medical ReasoningICD-Bench Difficulty Level 2 1.0
Accuracy85.63
2
Medical ReasoningICD-Bench Difficulty Level 1 1.0
Accuracy0.9675
2
Showing 6 of 6 rows