WaveFront Decoding: Parallelized Self-Speculative Decoding for Looped Language Models
Looped language models repeatedly apply a weight-shared block to increase effective depth without increasing parameter count, but the resulting T sequential recurrent-block calls per generated token substantially increase decoding latency. To address the issue, we introduce Wavefront Decoding (WFD), a training-free sel...
Hyeongju Ha, Jae-Joon Kim
· 0 citations