通过推测草稿树加速扩散采样
Accelerating Diffusion Sampling via Speculative Draft Trees
- Imperial College London（伦敦帝国理工学院）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
本文提出草稿树方法，通过将扩散模型推测采样与相对熵编码结合，丰富每轮候选状态并采用贪婪拒绝采样，在保证精确样本的同时，实现最高8.3%的加速。
中文摘要 AI 辅助
推测采样通过草拟廉价的候选状态并在保持目标分布精确的耦合下进行校正，从而加速扩散模型生成，减少了昂贵的目标评估次数。现有的扩散采样器，特别是基于反射最大耦合的采样器，受到拓扑约束：其前瞻草稿形成链图，即单一线性序列，这固有地限制了每次目标评估的接受率。我们将扩散模型中的推测采样与相对熵编码（REC）联系起来。这一视角表明前瞻不必是线性的，并激发了我们的核心贡献——草稿树，它丰富了每轮考虑的候选状态并降低了目标函数评估次数。我们进一步采用贪婪拒绝采样（一种REC算法）作为草稿-目标耦合，在保证精确目标样本的同时提高接受率。在多种目标模型和草稿模型上的实验表明，在实际设置中，与反射耦合基线相比，加速最高可达8.3%。
英文摘要
Speculative sampling accelerates diffusion model generation by drafting inexpensive candidate states and correcting them under a coupling that preserves the target distribution exactly, reducing the number of expensive target evaluations. Existing diffusion samplers, notably those based on reflection maximal coupling, are topologically constrained: their lookahead drafts form a chain graph, a single linear sequence, which inherently limits the acceptance rate per target evaluation. We connect speculative sampling in diffusion models to relative entropy coding (REC). This perspective shows the lookahead need not be linear and motivates our central contribution, draft trees, which enrich the candidates considered per round and lower the target function evaluations. We further adopt greedy rejection sampling, an REC algorithm, as the draft-target coupling, improving acceptance while guaranteeing exact target samples. Experiments across diverse target and draft models demonstrate up to 8.3% acceleration over the reflection coupling baseline in practical settings.
