Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Long-caption Cross-modal Retrieval on Urban-1k (test)
Loading...
91.1
T2I Recall@1
HyFL-CLIP
37.852
51.676
65.5
79.324
Jul 1, 2026
T2I Recall@1
I2T Recall@1
Updated 23d ago
Evaluation Results
Method
Method
Links
T2I Recall@1
I2T Recall@1
HyFL-CLIP
Model Type=Euclidean V...
2026.07
91.1
91.8
UNCHA + Ours
Model Type=State-of-th...
2026.07
77.1
76.8
HyCoCLIP + Ours
Model Type=State-of-th...
2026.07
75.3
76.7
MERU + Ours
Model Type=State-of-th...
2026.07
74.2
74.3
OpenCLIP
Model Type=Euclidean V...
2026.07
53.4
67.5
HyCoCLIP
Model Type=State-of-th...
2026.07
47.7
56.6
MERU
Model Type=State-of-th...
2026.07
45.7
51.9
UNCHA
Model Type=State-of-th...
2026.07
39.9
43.8
Feedback
Search any
task
Search any
task