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

Medical Question Answering on PubMedQA (FA, HS)

95.63Factual Accuracy (FA)

SPPFT

27.655645.302862.9580.5972Apr 14, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.04
95.6325.45
2026.04
95.528.65
2026.04
95.2453.1
2026.04
95.212.55
2026.04
95.1810.28
2026.04
94.527.24
2026.04
94.4111.63
2026.04
94.335.17
2026.04
94.2729.12
2026.04
94.1825.31
2026.04
93.8146.75
2026.04
93.4548.97
2026.04
93.2616.49
2026.04
93.229.38
2026.04
92.0527.4
2026.04
90.7215.39
2026.04
90.6222.37
2026.04
87.630.41
2026.04
85.3716.42
2026.04
83.459.54
2026.04
74.0515.37
2026.04
70.3712.49
2026.04
64.5360.72
2026.04
62.6745.25
2026.04
58.1520.31
2026.04
43.543.41
2026.04
43.0845.37
2026.04
30.2729.46