Fetching the paper…

CLIP-Lite: Information Efficient Visual Representation Learning with Language Supervision · Around