SigLIP B/16 + PrefixedIter Decoder (FT2)

Reported on 3 benchmarks across 1 task · 1 paper · 1 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision3 results

Zero-Shot Image ClassificationonOVIC Datasets (Wiki-H)
Prediction Score (mean of 3)· 2024-07-15
72.03
SOTA
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP Inversion arXiv:2407.11211
Zero-Shot Image ClassificationonOVIC Datasets (Val3K)
Prediction Score (mean of 3)· 2024-07-15
74.35
best: 76.5 (SigLIP B/16 + PrefixedIter Decoder (FT6))
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP Inversion arXiv:2407.11211
Zero-Shot Image ClassificationonOVIC Datasets (Val3K)
Top 1 Accuracy (mean of 3)· 2024-07-15
72.9
best: 75.04 (SigLIP B/16 + PrefixedIter Decoder (FT6))
Unconstrained Open Vocabulary Image Classification: Zero-Shot Transfer from Text to Image via CLIP Inversion arXiv:2407.11211