CLIP PRETRAINED MAMBA MODELS

Summary from the TLDR AI newsletter

The new Mamba model, trained using contrastive language-image pretraining (CLIP), shows impressive efficiency and performance in zero-shot image classification.