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

Medical Question Answering on PubMedQA Reasoning Required

82Accuracy

GPT-4 (Medprompt)

55.084862.072469.0676.0476Nov 28, 2023Apr 10, 2024Aug 22, 2024Jan 3, 2025May 17, 2025Sep 28, 2025Feb 9, 2026
Updated 2mo ago

Evaluation Results

MethodLinks
2023.11
82
2023.11
81.8
2023.11
79
2023.11
75.2
2026.02
60.31
2026.02
60.26
2026.02
58.8
2026.02
57.34
2026.02
56.96
2026.02
56.12