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

Multi-Discipline Reasoning on MMMU-Pro (pass@1, avg@8)

51.8Pass@1

ADHint

7.80819.22930.6542.071Dec 15, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
51.857.9
2025.12
46.557.9
2025.12
41.341.8
2025.12
40.440
2025.12
38.446
2025.12
35.332.3
2025.12
33.332
2025.12
30.730.1
2025.12
30.629.7
2025.12
30.228.7
2025.12
27.525.1
2025.12
2726.3
2025.12
25.838.3
2025.12
20.635.6
2025.12
20.323.1
2025.12
19.826
2025.12
18.730.9
2025.12
17.129.4
2025.12
9.519.3