D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding

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

D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding: 19 upvotes on Hugging Face Daily Papers, #12 of 38 papers on 2026-11-05. Day-by-day upvote history.

Upvotes over time

It had 17 upvotes by the next day, 19 upvotes by day 3, 19 upvotes by day 7, and has 19 now. It ranked #72 of 207 papers in its week and #110 of 340 in its month. Its code repository has 21 GitHub stars.

Target-conditioned diffusion drafters predict candidate tokens in parallel, yet each new block must wait for updated target-model hidden states. This forces drafting and verification to run sequentially, limiting speedup even though each block is generated in one pass. We introduce D3D, a standalone diffusion drafter that removes this inter-round dependency. D3D combines blockwise denoising with auxiliary autoregressive supervision to achieve long accepted prefixes without target hidden states. This independence lets D3D draft successor blocks during verification in one packed pass, enabling asynchronous speculative decoding. We progressively prune layers according to their effect on acceptance length to reduce the cost of packed drafting. Removing five of 36 layers retains over 99\% of mean acceptance length across three benchmarks with Qwen3-8B as the target. Across seven benchmarks, D3D averages 5.00times and 5.71times decoding speedup with Qwen3-8B and Qwen3-32B targets, compared with 4.03times and 3.67times for DFlash. On GSM8K with Qwen3-32B, speedup reaches 7.48times. Code is publicly available at https://github.com/fudan-generative-vision/D3D.

Paper page on Hugging Face · arXiv

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.