Consistency-guided Prompt Learning for Vision-Language Models

Benchmark Model Rank Results
prompt-engineering-on-food-101CoPrompt#2Harmonic mean: 91.40
prompt-engineering-on-sun397CoPrompt#4Harmonic mean: 81.31
prompt-engineering-on-oxford-iiit-pet-datasetCoPrompt#7Harmonic mean: 96.87
prompt-engineering-on-ucf101CoPrompt#8Harmonic mean: 83.07
prompt-engineering-on-imagenetCoPrompt#9Harmonic mean: 74.33
prompt-engineering-on-imagenet-rCoPrompt#11Top-1 accuracy %: 77.51
prompt-engineering-on-caltech-101CoPrompt#12Harmonic mean: 96.55
prompt-engineering-on-dtdCoPrompt#12Harmonic mean: 72.79
prompt-engineering-on-eurosatCoPrompt#12Harmonic mean: 85.84
prompt-engineering-on-imagenet-sCoPrompt#12Top-1 accuracy %: 49.43
prompt-engineering-on-oxford-102-flowerCoPrompt#14Harmonic mean: 85.71
prompt-engineering-on-fgvc-aircraftCoPrompt#15Harmonic mean: 39.76
prompt-engineering-on-imagenet-aCoPrompt#15Top-1 accuracy %: 50.50
prompt-engineering-on-stanford-carsCoPrompt#15Harmonic mean: 75.66