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

Multimodal Perception and Grounding on MME-RealWorld

51.94Score

SceneAlign

36.672840.636444.648.5636Jan 9, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2026.01
51.94
2026.01
50.33
2026.01
49.6
2026.01
48.86
2026.01
47.99
2026.01
47.36
2026.01
45.02
2026.01
44.45
2026.01
43.98
2026.01
40.42
2026.01
39.81
2026.01
39.42
2026.01
38.95
2026.01
38.32
2026.01
37.68
2026.01
37.68
2026.01
37.26