D3D 提出解耦式擴散草稿生成方法,支援非同步推測解碼
D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding
D3D 論文提出獨立擴散草稿器,將草稿生成與目標模型驗證解耦,支援非同步推測解碼。它在七個基準上的平均解碼加速比為 Qwen3-8B 的 5.00×、Qwen3-32B 的 5.71×,高於 DFlash 的 4.03×和 3.67×,並在 GSM8K 上達到最高 7.48×。
🚀 Introducing D3D: Decoupled Diffusion Drafting for Asynchronous Speculative Decoding!
Existing diffusion-based speculative drafters rely on target-model hidden states, forcing drafting and verification to execute sequentially.
D3D breaks this dependency with a standalone diffusion drafter, enabling asynchronous drafting and verification through packed speculative decoding.
Key highlights:
- Asynchronous speculation: Overlaps diffusion drafting with target verification, eliminating inter-round dependencies.
- Effective drafting: Combines blockwise denoising with auxiliary autoregressive supervision to achieve long accepted prefixes.
- Significant speedups: Achieves 5.00× and 5.71× average decoding speedups with Qwen3-8B and Qwen3-32B across seven benchmarks, outperforming DFlash (4.03× and 3.67×). Up to 7.48× on GSM8K!
📄 Paper: https://arxiv.org/abs/2611.06425
💻 Code: https://github.com/fudan-generative-vision/D3D
🤗 Models: https://huggingface.co/collections/fudan-generative-ai/d3d
We've released our code, model checkpoints, and benchmarking scripts.
We welcome feedback, questions, and discussions!
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co