AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Differentially Private Fine-Tuning of Diffusion Models

Yu-Lin Tsai, Yizhe Li, Chia-Mu Yu, Xuebin Ren, Po-Yu Chen, Zekai Chen, Francois Buet-Golfouse

ICCV 2025 · Federated learning and privacy › Differential privacy · approach: Private training with public data · 12 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 LoRA to QKV attention matrices (ablation)DP-LoRA · CelebA-32FID: 11.78 -> 7.71Improved
add LoRA to the output projection layer (ablation)DP-LoRA · CelebA-32FID: 9.82 -> 7.71Improved
add LoRA to the output projection layer (ablation)DP-LoRA · CelebA-64FID: 11.24 -> 8.41Improved
increase noise multiplicity from k=1 to k=4 (ablation)DP-LoRA · CelebA-32FID: 13.75 -> 7.71Improved
increase noise multiplicity from k=4 to k=8 (ablation)DP-LoRA · CelebA-64FID: 8.41 -> 8.94Hurt
increase noise multiplicity from k=4 to k=8 (ablation)DP-LoRA · CelebA-32FID: 7.71 -> 7.32Improved
raise LoRA rank from 16 to 64 (ablation)DP-LoRA · CelebA-32FID: 7.71 -> 10.16Hurt
lower LoRA rank from 16 to 8 (ablation)DP-LoRA · CelebA-32FID: 7.71 -> 8.09No clear effect
use parameter-efficient LoRA fine-tuning instead of DP-LDM full fine-tuningDP-LDM · CelebA-64 eps=10FID: 14.3 -> 8.4Improved
use LoRA latent diffusion instead of DP-LDMDP-LDM · CelebA-HQ eps=1FID: 25.6 -> 20.0Improved
use LoRA fine-tuning instead of DP-LDMDP-LDM · CIFAR-10 ResNet9 eps=10accuracy: 65.3 -> 73.98Improved
use LoRA latent diffusion instead of vanilla pixel-space DPDMDPDM · Fashion-MNIST eps=10accuracy: 86.2 -> 72.7Hurt

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