1 paper
Pouya Parsa, Raoof Zare Moayedi, Seongjin Choi
Prompt tuning adapts vision--language models with few trainable parameters, but existing approaches trade off efficiency and adaptation: static textual prompts can overfit source c…