Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Vision-Language Compositional Reasoning on What's-up
Loading...
44.2
Relation Score
CLIP-FT
40.352
41.351
42.35
43.349
Jun 11, 2026
Relation Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
Relation Score
CLIP-FT
Backbone=CLIP ViT-B/16
2026.06
44.2
MACCO-CLIP
Backbone=CLIP ViT-L/14
2026.06
42.4
CLIP-FT
Backbone=CLIP ViT-L/14
2026.06
42
CLIP
Backbone=CLIP ViT-B/16
2026.06
41.9
Random Chance
2026.06
41.7
MACCO-CLIP
Backbone=CLIP ViT-B/16
2026.06
41.7
SigLIP
Backbone=SigLIP ViT-B/16
2026.06
41.5
CLIP
Backbone=CLIP ViT-L/14
2026.06
41.2
MACCO-SigLIP
Backbone=SigLIP ViT-B/16
2026.06
40.8
SigLIP-FT
Backbone=SigLIP ViT-B/16
2026.06
40.5
Feedback
Search any
task
Search any
task