AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

An Adaptive Sampling Framework for Diffusion-based Dataset Distillation with High Fidelity and Diversity

Sunbeom Jeong, Sehwan Kim, Hyeonggeun Han, Hyungjun Joo, Sangwoo Hong, Jungwoo Lee

AAAI 2026 · Image generation › Fast diffusion sampling · approach: Distillation for few-step generation · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use adaptive sampling with repulsion regularization instead of D4Mdiffusion DD · ImageNet-1K IPC=50 (ResNet-18)accuracy (%): 55.2 -> 61.4Improved
use proposed method instead of D4Mdiffusion DD · ImageNet-1K IPC=10 (ResNet-18)accuracy (%): 27.9 -> 46.6Improved
use proposed method instead of D4Mdiffusion DD · ImageWoof IPC=50 (ResNet-18)accuracy (%): 71.6 -> 72.6Improved
use proposed method instead of D4Mdiffusion DD · ImageNet-100 IPC=10 (ResNet-18)accuracy (%): 30.2 -> 39.0Improved
use proposed method instead of D4M with ResNet-101diffusion DD · ImageNet-1K IPC=50 (ResNet-101)accuracy (%): 63.4 -> 67.8Improved
use proposed unaligned LDM instead of MGD3 aligned DiTdiffusion DD · ImageNet IPC=50accuracy (%): 60.2 -> 61.4Improved
use proposed method instead of Minimaxdiffusion DD · Synthetic image qualityFID: 57.9 -> 48.7Improved
use proposed method instead of Minimaxdiffusion DD · Synthetic image qualityPrecision: 0.88 -> 0.84Hurt
use proposed method instead of D4Mdiffusion DD · Synthetic image qualityRecall: 0.14 -> 0.45Improved
drop latent prototype generation (proposed vs D4M)diffusion DD · Distillation costGPU memory (GB): 6.4 -> 2.6Improved
training-free proposed method instead of Minimax fine-tuningdiffusion DD · Distillation costgeneration time per image (s): 10.1 -> 2.7Improved
use proposed method instead of RDED on ImageNette IPC=10 with ResNet-101diffusion DD · ImageNette IPC=10 (ResNet-101)accuracy (%): 54.0 -> 60.4Improved

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