Learning Hierarchical Prompt with Structured Linguistic Knowledge for Vision-Language Models

Benchmark Model Rank Results
prompt-engineering-on-caltech-101HPT#5Harmonic mean: 96.65
prompt-engineering-on-dtdHPT#7Harmonic mean: 72.16
prompt-engineering-on-eurosatHPT#8Harmonic mean: 84.82
prompt-engineering-on-fgvc-aircraftHPT#6Harmonic mean: 40.28
prompt-engineering-on-food-101HPT#11Harmonic mean: 91.01
prompt-engineering-on-imagenetHPT#7Harmonic mean: 74.17
prompt-engineering-on-imagenet-aHPT#6Top-1 accuracy %: 50.85
prompt-engineering-on-imagenet-rHPT#6Top-1 accuracy %: 77.38
prompt-engineering-on-imagenet-sHPT#4Top-1 accuracy %: 49.36
prompt-engineering-on-imagenet-v2HPT#2Top-1 accuracy %: 65.25
prompt-engineering-on-oxford-102-flowerHPT#2Harmonic mean: 87.16
prompt-engineering-on-oxford-iiit-pet-datasetHPT#5Harmonic mean: 96.71
prompt-engineering-on-stanford-cars-1HPT#9Harmonic mean: 75.57
prompt-engineering-on-sun397HPT#7Harmonic mean: 80.88
prompt-engineering-on-ucf101HPT#6Harmonic mean: 83.16