Few-Shot Adversarial Prompt Learning on Vision-Language Models | Read Paper on Bytez