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

AgentClinic

Benchmarks

Task NameDataset NameSOTA ResultTrend
Multimodal Clinical ReasoningAgentClinic_M offline (test)
Accuracy94.17
38
Clinical ReasoningAgentClinic_T offline (test)
Accuracy68.22
38
Medical DiagnosisAgentClinic OOD original (test)
Similarity (Sim)0.684
20
Interactive Medical DiagnosisAgentClinic MedQA
ICR53.9
12
Agentic medical interactionAgentClinic MedQA
Accuracy65.8
6
Showing 5 of 5 rows