Quantifying the failure of symmetry learning in linear diffusion models due to initialization and finite-sample effects
Paper recorded by Signals 4 on 2026-10-26 in cs.LG. Abstract reproduced from arXiv; link to the original below.
Published 2026-10-26 on arXiv · recorded by Signals 4 on 2026-10-27
Category: cs.LG · 机器学习 · first seen 2026-10-27
Abstract
Data distributions often possess one or more symmetries: for example, all orientations of any specific dog are included in the distribution of all possible dog pictures. Despite the omnipresence of such symmetries, modern generative modeling approaches may fail to incorporate or learn them, which can lead to lower sample efficiency and poorer generalization. To what extent do diffusion models fail
Read on arXiv →
Cite this page: Quantifying the failure of symmetry learning in linear diffusion models due to initialization and finite-sample effects: the #175 most recent of 749 cs.LG papers we have recorded (as of 2026-10-26). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/quantifying-the-failure-of-symmetry-learning-in-linear-diffusion-models-due-to-i.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable:
papers.json
Get 4 AI signals a day by email — free.
Get 4 AI signals a day by email — free