Unlocking Lossless Speedups in LLMs via Discrete Diffusion

Researchers introduced diffusion-augmented LLMs, a new class of models that combines autoregressive (AR) next-token prediction with discrete diffusion to enable parallel token generation, achieving lossless speedups in LLM inference. The approach defines an AR model distribution with a diffusion process, allowing faster generation without sacrificing quality.

Large Language Models LLMs owe much of their success to next-token prediction NTP , but their autoregressive AR structure requires slow, sequential token generation. To overcome this bottleneck, we introduce diffusion-augmented LLMs, a new class of models that defines an AR model distribution w