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

Multitask Language Understanding on MMLU physics

90.1MDR

A-Trust

-0.3823.1146.670.09Jun 3, 2025
Updated 3d ago

Evaluation Results

MethodLinks
2025.06
90.1
2025.06
89.3
2025.06
87.7
2025.06
85.4
2025.06
84.7
2025.06
84.3
2025.06
83.6
2025.06
82.7
2025.06
82.1
2025.06
80.5
2025.06
79.6
2025.06
77.5
2025.06
61.7
2025.06
60.8
2025.06
60.8
2025.06
60.7
2025.06
60.4
2025.06
59.3
2025.06
57.9
2025.06
57.7
2025.06
56.2
2025.06
55.3
2025.06
54.2
2025.06
53.8
2025.06
53.7
2025.06
53.1
2025.06
52.9
2025.06
51.8
2025.06
50.9
2025.06
50.5
2025.06
50.4
2025.06
50.3
2025.06
50.2
2025.06
48.9
2025.06
44.3
2025.06
43.1
2025.06
9.6
2025.06
9.5
2025.06
9.4
2025.06
8.2
2025.06
7.1
2025.06
5.7
2025.06
4.8
2025.06
4.5
2025.06
3.1