AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

PrivSynth: Alternating and Control-Based Optimization for Privacy and Utility in Synthetic Data

Xinyuan Zhao, Hanlin Gu, Guibao Song, Gongxi Zhu, Yifei Zou, Lixin Fan, Yuxing Han

CVPR 2026 · Federated learning and privacy › Differential privacy · approach: Differentially private synthetic data

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
apply two rounds of alternating optimization with PMP-based data selection (ablation)PrivSynth · DomainNetMIA ASR (%): 48.1% -> 1.9%Improved
apply two rounds of alternating optimization with PMP-based data selection (ablation)PrivSynth · DomainNet1-LPIPS: 0.67 -> 0.49Improved
apply two rounds of alternating optimization (ablation)PrivSynth · DomainNetdownstream accuracy: remains above 67%No clear effect
directly filter out privacy-sensitive data instead of utility-aware selection (ablation)PrivSynth · downstream classificationutilityHurt
increase ratio of selected synthetic data (ablation)PrivSynth · downstream classificationaccuracy and privacy leakageMixed
increase privacy weight phi1 (ablation)PrivSynth · membership leakageMIA ASRImproved
increase privacy weight phi2 (ablation)PrivSynth · pixel-level leakageLPIPSImproved
use PrivSynth selection instead of DP, Mask, Dedup on synthetic datasynthetic data generation · DomainNet and ImageNetFID and KIDImproved
select data with classification loss as proxy for unknown downstream taskPrivSynth · ImageNet object detectiondetection error (1-AP50)Improved

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