BEiTModel

E1312485 UNEXPLORED

BEiTModel is a vision transformer-based image encoder that uses a BERT-style masked image modeling pretraining strategy for various computer vision tasks.

All labels observed (1)

Label Occurrences
BEiTModel canonical 1

How this entity was disambiguated

Referenced by (1)

Full triples — surface form annotated when it differs from this entity's canonical label.