Variational Adapter for Cross-modal Similarity Representation
Open paper
Benchmark
Model
Rank
Results
prompt-engineering-on-imagenet-a
VACSR
–
Top-1 accuracy %: 52.4
prompt-engineering-on-imagenet-r
VACSR
–
Top-1 accuracy %: 78.4
prompt-engineering-on-imagenet-v2
VACSR
–
Top-1 accuracy %: 65.7
Rank counts only results with a code link.