Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Concept Detection on CLEVR
Loading...
99.2
F1 Score
SuperAct
48.448
61.624
74.8
87.976
Dec 4, 2025
F1 Score
Updated 4d ago
Evaluation Results
Method
Method
Links
F1 Score
SuperAct
Backbone=CLIP (92% depth)
2025.12
99.2
CLS
Backbone=CLIP (92% depth)
2025.12
89.8
MeanTok
Backbone=CLIP (92% depth)
2025.12
60.9
RandTok
Backbone=CLIP (92% depth)
2025.12
50.4
LastTok
Backbone=CLIP (92% depth)
2025.12
50.4
Feedback
Search any
task
Search any
task
