Reproduce “Equal Path Cost, Unequal Output Effects: Understanding Perturbation Propagation in Diffusion Models” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Wei Guo, Yaowen Zhang, Xingtong Ge, Jun Zhang
The authors analyze how perturbations propagate through a diffusion sampler and affect its output distribution. Their decomposition separates trajectory-level KL into output KL and nonnegative bridge cost, while analysis tracks propagation. In a caching test across four models, propagation-based rankings achieve Spearman correlations of 0.9524 to 0.9762; this supports ranking tested windows, not exact error prediction or speedup.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.