CChinmay Bansal·Jul 19How Vision-Language Models Actually 'See'The not-so-boring guide to CLIP, SigLIP, and why AI can find your cat photos without ever being told what a cat is10 min read·chinmaybansal.hashnode.dev
RRitwik Raha·Aug 2, 2024Choosing Between SigLIP and CLIP for Language Image PretrainingUnderstand the changes from CLIP to SigLIP in vision-language models, highlighting the advantages of contrastive learning and how they enhances efficiency11 min read·blog.ritwikraha.dev