AI Paper Atlas

Research map / Federated learning and privacy / Personalized and heterogeneous FL

GDFA: Geometry-Driven Federated Unlearning with Directional Task Vector Alignment

Xiuting Weng, Ruizhi Pu, Yuanhang Yao, Kun Yue, Zhiwen Tang, Lixing Yu

CVPR 2026 · Federated learning and privacy › Personalized and heterogeneous FL · approach: Federated learning with pretrained models

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
negate the merged task vector with larger magnitude (lambda from 0 to -1) (ablation)GDFA · FMNIST (IID)FA: 98.21 -> 1.45Improved
migrate global model to flat region via perturbation radius rho = 0.5 (ablation)GDFA · CIFAR10 (Dir 1)RA / FAImproved
use too small perturbation radius rho (ablation)GDFA · CIFAR10 (Dir 1)RA / FAHurt
use too large perturbation radius rho (ablation)GDFA · CIFAR10 (Dir 1)RAHurt
increase number of perturbed models K (use K = 5) (ablation)GDFA · CIFAR10 (Dir 1)RA / timeMixed
flat-region migration plus sign-consensus task vectors vs SIFU on retained accuracyGDFA · CIFAR10 (Dir 0.5)RA: 55.17 -> 69.33Improved
GDFA versus retraining on unlearning timeGDFA · CIFAR10 (Dir 0.5)time (s): 1161.33 -> 31.92Improved
GDFA versus Eraser on forgotten accuracyGDFA · CIFAR10 (Dir 0.5)FA: 0.68 -> 1.57Hurt
GDFA versus Retrain on IID CIFAR10 forgotten accuracyGDFA · CIFAR10 (IID)FA: 2.32 -> 3.04Hurt

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