<ul><li>MaCP is a new adaptation method, Minimal yet Mighty adaptive Cosine Projection, designed for fine-tuning large foundation models.</li><li>It utilizes cosine projection to improve model efficiency and accuracy by projecting weight changes into discrete cosine space.</li><li>MaCP has been shown to be effective across various tasks including natural language processing, image classification, and video understanding.</li><li>Experiments demonstrate that MaCP offers superior accuracy, reduced computational complexity, and lower memory requirements compared to existing methods.</li></ul>

MaCP: Minimal yet Mighty Adaptation via Hierarchical Cosine Projection

Discover more