← Back to all news
arXiv AI By Wei Guo, Yaowen Zhang, Xingtong Ge, Jun Zhang

Equal Path Cost, Unequal Output Effects: Understanding Perturbation Propagation in Diffusion Models

Read the original on arXiv AI →

The paper introduces a theoretical framework to study how perturbations introduced during diffusion model sampling affect the generated output. It defines a path cost based on the Kullback–Leibler divergence between perturbed and reference trajectories, showing that this cost bounds but does not fully determine output changes. Using a response identity, the authors link local perturbation propagation to changes in feature means, revealing that first‑order responses can miss significant output variations. Experiments on pretrained diffusion models demonstrate varying output sensitivity across sampling stages and spatial frequencies, and the framework successfully identifies problematic sampling intervals in cache‑based acceleration.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

Hugging Face Trending Papers
Oct 8

Equal Path Cost, Unequal Output Effects: Understanding Perturbation Propagation in Diffusion Models

The paper investigates how perturbations introduced during the sampling process of diffusion models influence the final generated output. It introduces a theoretical framework that quantifies perturbation strength via the Kullback–Leibler divergence, called path cost, and shows that this metric bounds but does not fully determine output changes. Using this framework, the authors derive a response identity linking local perturbations to feature mean changes, and validate their theory through experiments on pretrained diffusion models and cache‑based acceleration techniques.

diffusion
More like this →
arXiv Machine Learning
Sep 22

Leveraging Inference-Time Compute for Diffusion Models via Global Scheduling of Denoising Trajectories

The paper studies how to allocate a fixed computational budget across the denoising steps of diffusion models to improve sample quality at deployment. It shows that the expected benefit of evaluating multiple candidates at a step can be decomposed into a step‑specific sensitivity and a universal sample‑size factor, and that the optimal allocation follows a water‑filling structure. Experiments demonstrate that this allocation achieves the same quality as a uniform strategy while reducing function evaluations by 20–50%.

By Yuan Cao, Yifu Tang, Hangqi Li, Zeyu Zheng
diffusionbenchmarks
More like this →