Publications

When Does Visual Prompting Outperform Linear Probing for Vision-Language Models? A Likelihood Perspective