AI Paper Atlas

Research map / Federated learning and privacy / Differential privacy

When Privacy Meets Recovery: The Overlooked Half of Surrogate-Driven Privacy Preservation for MLLM Editing

Siyuan Xu, Yibing Liu, Peilin Chen, Yung-Hui Li, Shiqi Wang, Sam Kwong

AAAI 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
train recovery model SOER instead of VISII, Prompt-Diffusion or Edit-Transfersurrogate-to-original recovery · SPPE test, TextualSSIM: 0.7483 -> 0.8049 (vs E-tran)Improved
use SOER on unseen images with full-image surrogateSOER · InstructPix2Pix subsetDirS: 0.1717 -> 0.2077 (vs E-tran)Improved
SOER on source integrity vs baselinesSOER · SPPE Textual source integritySSIM: VISII 0.689 vs SOER 0.636 (GT 0.627)Mixed
add privacy mask guidance (ablation)SOER · SPPESSIM: 0.789 -> 0.802Improved
add edit map guidance (ablation)SOER · SPPESSIM: 0.787 -> 0.802Improved
add text prompt guidance (ablation)SOER · SPPESSIM: 0.714 -> 0.802Improved
use VAE-based transformer encoder instead of CNN for mask/edit maps (ablation)SOER · SPPEDirS: 0.592 -> 0.666Improved
use transformer-based spatial encoder instead of single CLS token (ablation)SOER · SPPEDirS: 0.620 -> 0.666Improved

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