MATH · IN · MODELS

BEiT (BERT Pre-Training of Image Transformers)

Microsoft

Structures found in this family (1)

By model (1)

BEiT-Large (ImageNet-22k pretrained)

Papers

Linearly Mapping from Image to Text Space (2023)