AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

CoDA: From Text-to-Image Diffusion Models to Training-Free Dataset Distillation

Letian Zhou, Songhua Liu, Xinchao Wang

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 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
add density-based core distribution discovery and alignment guidanceSDXL · ImageIDC IPC10accuracy: 40.1 -> 58.5Improved
CoDA vs prior SOTA (DiT+MGD3) on ImageIDC IPC50DiT+MGD3 · ImageIDC IPC50accuracy: 72.1 -> 77.6Improved
CoDA vs prior SOTA on ImageWoof at low IPCMGD3 / Minimax · ImageWoof IPC10-20 (ResNet-18)accuracy: IPC20: 42.5 (Minimax) vs 42.0Mixed
CoDA vs prior SOTA on ImageWoof at high IPCMGD3 · ImageWoof IPC100 (ResNet-18)accuracy: 68.8 -> 71.4Improved
CoDA vs IGD (ImageNet-trained model) at IPC10IGD · ImageNet-1K IPC10accuracy: 46.2 -> 44.3Hurt
CoDA vs IGD at IPC50IGD · ImageNet-1K IPC50accuracy: 60.3 -> 60.4Improved
CoDA vs D4M, an off-the-shelf competitorD4M · ImageNet-1K IPC10accuracy: 27.9 -> 44.3Improved
apply CoDA to Places365 with no modificationSDXL · Places365 10 classes IPC10accuracy: 33.9 -> 47.0Improved
use K-Means on SDXL instead of nothing (ablation)SDXL · Places365 10 classes IPC10accuracy: 33.9 -> 34.4No clear effect
add alignment stage (generated set G over real set R) (ablation)CoDA discovery output · ImageIDC IPC50accuracy: 75.7 -> 77.6Improved
replace K-Means with CoDA discovery in UMAP-VAE space (ablation)prototype selection · ImageIDC IPC10accuracy: 54.8 -> 58.5Improved
add UMAP preprocessing before clustering (ablation)CoDA · ImageIDC IPC10accuracy: 45.8 -> 58.5Improved

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