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

Spatial Reasoning on VG-B SpatialScore-Hard

43.8Accuracy

TVP (Generalize)

4.69614.8482535.152Dec 24, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
43.8
2025.12
39.1
2025.12
20.3
2025.12
12.5
2025.12
9.4
2025.12
7.8
2025.12
7.8
2025.12
6.2