AI RESEARCH
CLIP-SVD: Efficient and Interpretable Vision-Language Adaptation via Singular Values
arXiv CS.CL
•
ArXi:2509.03740v3 Announce Type: replace-cross Vision-language models (VLMs) like CLIP have shown impressive zero-shot and few-shot learning capabilities across diverse applications. However, adapting these models to new fine-grained domains remains difficult due to reliance on prompt engineering and the high cost of full model fine-tuning. Existing adaptation approaches rely on augmented components, such as prompt tokens and adapter modules, which could limit adaptation quality, destabilize the model, and compromise the rich knowledge learned during pre