AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

Reinforcement-Guided Synthetic Data Generation for Privacy-Sensitive Identity Recognition

Xuemei Jia, Jiawei Du, Hui Wei, Jun Chen, Joey Tianyi Zhou, Zheng Wang

CVPR 2026 · Federated learning and privacy › Differential privacy · approach: Differentially private synthetic data · 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 Dynamic Sample Selection (DSS) to Base-DiT (ablation)Base-DiT · CASIA-WebFace subset face verificationavg accuracy: gain of 2.2%Improved
add semantic reward (SC) (ablation)Base-DiT + DSS · CASIA-WebFace subset face verificationavg accuracy: 78.18%Improved
add coverage reward (DC) (ablation)Base-DiT + DSS + SC · CASIA-WebFace subset face verificationavg accuracy: 78.18% -> 78.65%Improved
add expression reward (ED) (ablation)Base-DiT + DSS + SC + DC · CASIA-WebFace subset face verificationavg accuracy: 78.65% -> 79.07%Improved
reinforcement-guided DiT synthetic augmentation vs baselineResNet-50 · Market-1501mAP: 85.4 -> 88.6 (3.2% gain)Improved
reinforcement-guided DiT synthetic augmentation vs baselineResNet-50 · CUHK03-NPmAP: 76.6% (2.5% gain)Improved
use text-guided synthetic data from GIF-SDResNet-50 · Market-1501mAP: mAP decrease of 10.5% vs. baselineHurt
use simulation-based augmentation (FineGPR, InfinitePerson)ResNet-50 · Market-1501mAP: 82.4 and 57.3 vs 85.4Hurt
real-image augmentation (R-Erasing, CIDAM, CaAug)ResNet-50 · Market-1501mAP: about 2% mAP gainImproved
our method vs NegFaceDiffface recognition ResNet-50 + CosFace · CASIA-WebFace subset face verificationavg accuracy: surpassing by 0.94%Improved
our synthetic data vs prior methods on Indian and Asian subsetsface recognition · RFWaccuracy: Indian 1.0%, Asian 0.9%Improved
use ViT backbone with our methodViT-16 · Market-1501 / CUHK03-NPrank-1 / mAP: clear Rank-1 boosts, mAP comparable to R-ErasingMixed

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