AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Do We Need All the Synthetic Data? Targeted Image Augmentation via Diffusion Models

Dang Nguyen, Jiping Li, Jinghao Zheng, Baharan Mirzasoleiman

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Distributional control of diffusion · 2 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
augment only slow-learnable examples with TADA synthetic images instead of USEFUL upsamplingConvNeXt-T on CIFAR-10 (SGD) · CIFAR-10test error (%): 34.16 -> 27.40Improved
augment only slow-learnable examples with TADA instead of USEFUL upsamplingSwin-T on CIFAR-10 (SGD) · CIFAR-10test error (%): 14.93 -> 14.57Improved
add TADA to DiffuseMix when fine-tuning pretrained ResNet18DiffuseMix · Flowers-102test error (%): 8.92 -> 8.08Improved
add TADA to DiffuseMix when fine-tuning pretrained ResNet18DiffuseMix · Aircrafttest error (%): 25.65 -> 25.12Improved
augment all data with DiffuseMix versus no augmentationpretrained ResNet18 · Flowers-102test error (%): 8.55 -> 8.92Hurt
augment only the 30-40% slow-learnable examples instead of all examples (Syn-all) or random subset (Syn-rand) (ablation)TADA, ResNet18 · CIFAR-10 and CIFAR-100test errorImproved
increase amplification factor k above 2 with upsampling of real images (ablation)USEFUL, ResNet18 CIFAR-10 · CIFAR-10test errorHurt
increase amplification factor k to 5 with synthetic generation (ablation)TADA, ResNet18 · CIFAR-10, CIFAR-100test error: best k=5 (k=4 on TinyImageNet)Improved
generate synthetic images from noised real images instead of random noise (ablation)TADA, ResNet18 · CIFAR-10test errorImproved
use 50 denoising steps instead of fewer or many more (ablation)TADA with SGD and SAM · CIFAR-10test errorMixed
select slow-learnable examples by clustering model outputs instead of high loss or misclassification (ablation)TADA · appendix selection ablationtest errorImproved
apply TrivialAugment to targeted subset instead of full data (ablation)upsampled data · CIFAR-10test errorImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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