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

Medical Knowledge Evaluation on MMedbench English subset (val)

60.33Accuracy

LLaMA3-8B

20.570830.892941.21551.5371May 30, 2024
Updated 1mo ago

Evaluation Results

MethodLinks
2024.05
60.33
2024.05
60.02
2024.05
59.07
2024.05
58.68
2024.05
58.68
2024.05
58.44
2024.05
57.97
2024.05
57.5
2024.05
55.3
2024.05
46.9
2024.05
46.74
2024.05
46.58
2024.05
46.03
2024.05
45.56
2024.05
45.09
2024.05
44.85
2024.05
38.73
2024.05
38.33
2024.05
38.1
2024.05
38.02
2024.05
37.86
2024.05
37.55
2024.05
37.23
2024.05
36.68
2024.05
36.53
2024.05
34.64
2024.05
34.33
2024.05
34.33
2024.05
33.78
2024.05
32.21
2024.05
29.9
2024.05
27.65
2024.05
27.57
2024.05
27.49
2024.05
27.42
2024.05
22.1