Prune Once: Retraining-Free Task-Agnostic Pruning for Vision-Language Models

Open paper
Benchmark Model Rank Results
image-classification-on-cifar-10PORTA (OpenAI CLIP-ViT-bigG, 55% sparsity)#155Percentage correct: 95.58
image-classification-on-cifar-100PORTA (OpenAI CLIP-ViT-bigG, 55% sparsity)#160Percentage correct: 80.27

Rank counts only results with a code link.