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

Multi-modal Evaluation on MME-RW

31.9Mean Accuracy

TTAug

25.97227.51129.0530.589Oct 3, 2025
Updated 14d ago

Evaluation Results

MethodLinks
2025.10
31.9---
2025.10
31.4---
2025.10
31.1---
2025.10
30.9---
2025.10
27.8---
2025.10
27.8---
2025.10
27.6---
2025.10
27.6---
2025.10
26.4---
2025.10
26.2---
2026.02
-47.21--
2026.02
-49.7--
2026.02
-49.5--
2026.02
-50.9--
2026.02
-51.49--
2025.09
-45.246.442.3
2025.09
-61.464.340.1
2025.09
-62.365.142
2025.09
-63.866.742.9