Block Transformer: Global-to-Local Language Modeling for Fast Inference

Devs

Block Transformer: Global-to-Local Language Modeling for Fast Inference | Read Paper on Bytez