1 paper
Jian Yang, Shawn Guo, Wei Zhang +16
Looped Transformers scale latent computation by repeatedly applying shared blocks, but sequential looping increases latency and KV-cache memory with the loop count. Parallel loop T…