paper-with-me

홈 › Papers

Diffusion Models are Secretly Exchangeable: Parallelizing DDPMs via Autospeculation

2025-05-06 · Hengyuan Hu, Aniket Das, Dorsa Sadigh, Nima Anari

Denoising Diffusion Probabilistic Models (DDPMs) have emerged as powerful tools for generative modeling. However, their sequential computation requirements lead to significant inference-time bottlenecks. In this work, we utilize the connection between DDPMs and Stochastic Localization to prove that, under an appropriate reparametrization, the increments of DDPM satisfy an exchangeability property. This general insight enables near-black-box adaptation of various performance optimization techniques from autoregressive models to the diffusion setting. To demonstrate this, we introduce \emph{Autospeculative Decoding} (ASD), an extension of the widely used speculative decoding algorithm to DDPMs that does not require any auxiliary draft models. Our theoretical analysis shows that ASD achieves a $\tilde{O} (K^{\frac{1}{3}})$ parallel runtime speedup over the $K$ step sequential DDPM. We also demonstrate that a practical implementation of autospeculative decoding accelerates DDPM inference significantly in various domains.

📄 PDF Abstract BibTeX arXiv:2505.03983

Code (0)

등록된 구현이 없습니다.

Tasks

Denoising

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

Fast Diffusion Probabilistic Model Sampling through the lens of Backward Error Analysis

2023-04-22 · Yansong Gao, Zhihong Pan, Xin Zhou, Le Kang 외

Denoising diffusion probabilistic models (DDPMs) are a class of powerful generative models. The past few years have witnessed the great success of DDPMs in generating high-fidelity samples. A significant limitation of th…

Denoising

An overview of diffusion models for generative artificial intelligence

2024-12-02 · Davide Gallon, Arnulf Jentzen, Philippe von Wurstemberger

This article provides a mathematically rigorous introduction to denoising diffusion probabilistic models (DDPMs), sometimes also referred to as diffusion probabilistic models or diffusion models, for generative artificia…

Denoising

Fingerprinting Denoising Diffusion Probabilistic Models

2025-01-01 · CVPR 2025 1 · Huan Teng, Yuhui Quan, Chengyu Wang, Jun Huang 외

Diffusion models, especially denoising diffusion probabilistic models (DDPMs), are prevalent tools in generative AI, making their intellectual property (IP) protection increasingly important. Most existing IP protect…

Denoising

ShiftDDPMs: Exploring Conditional Diffusion Models by Shifting Diffusion Trajectories

2023-02-05 · Zijian Zhang, Zhou Zhao, Jun Yu, Qi Tian

Diffusion models have recently exhibited remarkable abilities to synthesize striking image samples since the introduction of denoising diffusion probabilistic models (DDPMs). Their key idea is to disrupt images into nois…

DenoisingImage Generation

Parallel Sampling via Autospeculation

2025-11-11 · Nima Anari, Carlo Baronio, CJ Chen, Alireza Haqi 외 arxiv

We present parallel algorithms to accelerate sampling via counting in two settings: any-order autoregressive models and denoising diffusion models. An any-order autoregressive model accesses a target distribution $μ$ on …