Signals 4 · free daily AI digest

Unifying Distributional Training for One-Step Visual Generation

Paper recorded by Signals 4 on 2026-09-28 in cs.LG. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-28 on arXiv · recorded by Signals 4 on 2026-09-29

Category: cs.LG · 机器学习 · first seen 2026-09-29

Abstract

\emph{Distributional training} provides collective supervision for one-step visual generation by matching real and generated features in frozen representation spaces. We introduce \emph{a unified theoretical framework} that separates distribution modeling from matching discrepancy and connects global objectives to pointwise feature updates through Wasserstein gradient flow. Under this framework, F

Read on arXiv →

#2 most recent of 322 cs.LG papers we have recorded · ↑ newer: PDMD: Projected Distribution Matching Distillation for Video Diffusion · ↓ older: Statistical Learning of Contractive Dynamical Representations for Comp
Cite this page: Unifying Distributional Training for One-Step Visual Generation: the #2 most recent of 322 cs.LG papers we have recorded (as of 2026-09-28). Source: Signals 4 (Signals API) — https://data.jiangzhang.ca/signals4/t/papers/unifying-distributional-training-for-one-step-visual-generation.html
Free to quote with attribution to “Signals 4 (Signals API)”. Machine-readable: papers.json
Related: More cs.LG papers · arXiv signals · All papers · Today in AI
Get 4 AI signals a day by email — free.
Subscribe free → See all plans →
Get 4 AI signals a day by email — free
All models · All repos · By company · Daily editions