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

Visual Reasoning on MMMU-Pro (Discriminability)

19.5Discriminability

Simulatability

-0.3644.7939.9515.107Sep 30, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.09
19.5
2025.09
16.9
2025.09
15.4
2025.09
14.6
2025.09
14.3
2025.09
12.7
2025.09
12.4
2025.09
10.8
2025.09
10.1
2025.09
9.6
2025.09
9.6
2025.09
9
2025.09
9
2025.09
8.2
2025.09
7.9
2025.09
5.9
2025.09
4.7
2025.09
3.8
2025.09
1.2
2025.09
1.1
2025.09
0.8
2025.09
0.6
2025.09
0.6
2025.09
0.4