arXiv:2410. 06814v2 Announce Type: replace Abstract: Over-parameterized models are typically vulnerable to membership inference attacks, which aim to determine whether a specific sample is included in the training of a given model.
arXiv:2608.28934v1 Announce Type: new Abstract: Differential privacy (DP) has traditionally been used to provide theoretical upper bounds on an algorithm's stability to changing its training data. In...
arXiv:2602. 02819v4 Announce Type: replace Abstract: Membership Inference Attacks (MIAs) aim to distinguish training points (members) from unseen data (non-members), and are widely used to quantify memorization and assess privacy risks.
The paper investigates whether inexpensive spectral metrics from the heavy‑tailed self‑regularisation framework can predict membership inference attack (MIA) vulnerability, offering a scalable alternative to costly shadow‑model attacks. Experiments on image and tabular classification tasks show that stable rank correlates positively with overall MIA success, while Log alpha‑Norm correlates negatively with MIA risk in low false‑positive regimes, outperforming conventional generalisation gap measures. These findings suggest that neural network spectra contain privacy leakage signals not captured by traditional overfitting metrics, pointing to spectral analysis as a promising direction for privacy auditing.
arXiv:2607. 02903v1 Announce Type: cross Abstract: Explainability is central to building trustworthy AI, yet explanation interfaces can inadvertently provide adversaries with an expanded privacy-related attack surfaces.
Adaptive Diffusion Freezing (ADF) is a new privacy‑preserving framework for diffusion models that protects against membership inference attacks (MIAs). It uses cross‑timestep adaptive freezing training, where a mask matrix controls which data subsets participate at each diffusion timestep, reducing over‑memorization and aligning model behavior for member and non‑member samples. A pretraining‑based risk‑aware freezing policy estimates MIA risk and suppresses high‑risk subset‑timestep pairs, achieving a superior privacy‑utility‑efficiency trade‑off across multiple datasets.
arXiv:2605. 27292v2 Announce Type: replace Abstract: Privacy auditing aims to empirically assess privacy leakage in machine learning models using membership inference attacks (MIAs), and to derive lower bounds on differential privacy (DP) parameters.
The paper introduces a general learning framework that protects privacy in federated learning by distorting model parameters, enabling a trade‑off between privacy and utility. The algorithm supports arbitrary privacy measurements and delivers personalized utility‑privacy balances for each parameter, client, and communication round. The authors prove that the gap between their algorithm’s utility loss and the optimal loss is sub‑linear in iterations, provide a convergence rate, and demonstrate empirically that their method outperforms baselines under the same privacy budget.
arXiv:2608. 13773v1 Announce Type: cross Abstract: Neural networks are increasingly deployed in high-stakes applications with growing privacy leakage concerns.
Batch Normalization (BN) is widely used to speed up and stabilize deep neural network training, yet its effect on privacy and memorization has been largely unexplored. This study shows that BN significantly increases the memorization of atypical or outlier samples, as evidenced by unintended memorization, per-sample influence, and heightened susceptibility to membership inference attacks across multiple datasets and architectures. A mechanistic analysis of the BN backward pass reveals that BN amplifies the per‑step margin growth of outlier samples during training, thereby intensifying their influence.
arXiv:2606. 09401v1 Announce Type: new Abstract: Recent work has applied differential privacy (DP) to adapt large language models (LLMs) for sensitive applications, offering theoretical guarantees.
arXiv:2606. 17464v1 Announce Type: new Abstract: Membership inference attacks (MIAs) are a canonical way to assess a machine learning model's privacy properties.
arXiv:2606. 17110v1 Announce Type: cross Abstract: Large Language Models are increasingly trained on proprietary or sensitive data, from private healthcare and financial records to user conversations containing secrets.