AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Efficient Hybrid Distillation: Synergizing Score and Adversarial Objectives for One-Step Diffusion

Fei Peng, Junqiang Wu, Haoxian Tan, Jun Zhou, Jie Hu, Xiaoming Wei, Jie Guo, Xiu Li

NeurIPS 2026 · Image generation › Fast diffusion sampling · approach: Fast diffusion ODE solvers · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes a hybrid distillation approach that synergizes score-based and adversarial objectives to produce one-step diffusion generators.

Problem

Accelerating diffusion models via distillation into one-step generators; existing hybrid methods combine distillation objectives simply, leaving effective synergy underexplored.

Approach

Focuses on how to effectively synergize score distillation and adversarial objectives rather than simply summing them, for one-step diffusion distillation.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Regime-Adaptive Sampling (score high-noise, adversarial low-noise) instead of uniform (ablation)EASD, SiT-XL/2+REPA · ImageNet 256x256FID: 1.60 -> 1.47Improved
hard truncation instead of uniform timestep sampling (ablation)EASD · ImageNet 256x256FID: 1.60 -> 1.51Improved
soft Logit-Normal biasing instead of hard truncation (ablation)EASD · ImageNet 256x256FID: 1.51 -> 1.47Improved
add Regime-Adaptive Sampling to DMD2 (ablation)DMD2 · ImageNet 64x64FID@100K / FID@200K: 1.72 -> 1.50; 1.52 -> 1.35Improved
replace fake-score-updating adversarial step with frozen-teacher discriminator (ablation)DMD2 with EASD · ImageNet 64x64FID@100K: 1.50 -> 1.46; 28h -> 23hImproved
bias score distillation to low noise (ablation)Score distillation only · ImageNet 256x256FID: 2.14 -> 2.99Hurt
bias adversarial distillation to high noise (ablation)Adversarial distillation only · ImageNet 256x256FID: 2.31 -> 2.80Hurt
attach discriminator to frozen teacher instead of updating fake score (ablation)EASD · ImageNet 256x256FID: 1.85 -> 1.47Improved
attach discriminator at middle block instead of shallow or deep (ablation)EASD · ImageNet 256x256FID: 1.55 (block 7), 1.70 (block 21) -> 1.47Improved
add Fake Score EMA to uniform-sampling baseline (ablation)EASD baseline · ImageNet 256x256FID: 1.60 -> 1.56Improved
add Iterative Warmup (ablation)EASD · ImageNet 256x256FID: 1.41 -> 1.39 (60 epochs)Improved
add Latent Mode Seeking loss (ablation)EASD baseline · ImageNet 256x256FID: 1.60 -> 1.58Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy