Skip to main content

D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding

Baoyou Chen, Ruiqiao Mei, Guancong Lin, Tao Song, Weihao Yuan, Qun Liu, Siyu Zhu

Trains a standalone diffusion model to draft several tokens at once, then prunes it using verification acceptance as the criterion. Prepares continuation branches while the target model checks the current draft, reusing a branch when its prefix matches.

First public: · Added to this list: November 9, 2026

Cite this paper

ArXiv preprint

arXiv:2611.06425 · Initial submission: 2026-11-04

BibTeX · CSL-JSON

Explore this approach

Alternative generation methods — Change how intermediate steps are generated, revised, or searched—for example, by refining several steps in parallel. Some produce written chains of thought.