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

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

64.4pass@1

ADHint

16.45628.90341.3553.797Dec 15, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
64.470.1
2025.12
6171.7
2025.12
56.456.4
2025.12
55.355.6
2025.12
51.750
2025.12
51.450
2025.12
47.549.1
2025.12
47.255.5
2025.12
44.647.4
2025.12
44.443
2025.12
44.442.6
2025.12
40.639.3
2025.12
35.954.2
2025.12
34.344.6
2025.12
31.251
2025.12
30.647.7
2025.12
28.646.3
2025.12
18.627.9
2025.12
18.335.7