SelectiveShield: Lightweight Hybrid Defense against Gradient Leakage in Federated Learning

Federated Learning FL enables collaborative model training on decentralized data but remains vulnerable to gradient leakage attacks that can reconstruct sensitive user information. Existing defense mechanisms, such as differential privacy DP and homomorphic encryption HE, often introduce a...

Benchmarking Fraud Detectors on Private Graph Data

We introduce the novel problem of benchmarking fraud detectors on private graph-structured data. Currently, many types of fraud are managed in part by automated detection algorithms that operate over graphs. We consider the scenario where a data holder wishes to outsource development of fraud...

Multi-Stage Prompt Inference Attacks on Enterprise LLM Systems

Large Language Models LLMs deployed in enterprise settings e.g., as Microsoft 365 Copilot face novel security challenges. One critical threat is prompt inference attacks: adversaries chain together seemingly benign prompts to gradually extract confidential data. In this paper, we present a...

Optimizing Canaries for Privacy Auditing with Metagradient Descent

In this work we study black-box privacy auditing, where the goal is to lower bound the privacy parameter of a differentially private learning algorithm using only the algorithm's outputs i.e., final trained model. For DP-SGD the most successful method for training differentially private deep...

A Privacy-Preserving Framework for Advertising Personalization Incorporating Federated Learning and Differential Privacy

To mitigate privacy leakage and performance issues in personalized advertising, this paper proposes a framework that integrates federated learning and differential privacy. The system combines distributed feature extraction, dynamic privacy budget allocation, and robust model aggregation to balan...

A Review of Privacy Metrics for Privacy-Preserving Synthetic Data Generation

Privacy Preserving Synthetic Data Generation PP-SDG has emerged to produce synthetic datasets from personal data while maintaining privacy and utility. Differential privacy DP is the property of a PP-SDG mechanism that establishes how protected individuals are when sharing their sensitive data. I...

Differentially Private Federated Low Rank Adaptation beyond Fixed-Matrix

Large language models LLMs typically require fine-tuning for domain-specific tasks, and LoRA offers a computationally efficient approach by training low-rank adapters. LoRA is also communication-efficient for federated LLMs when multiple users collaboratively fine-tune a global LLM model without...

Crypto-Assisted Graph Degree Sequence Release under Local Differential Privacy

Whitepaper called Crypto-Assisted Graph Degree Sequence Release Under Local Differential Privacy...

Optimal Debiased Inference on Privatized Data Via Indirect Estimation and Parametric Bootstrap

We design a debiased parametric bootstrap framework for statistical inference from differentially private data. Existing usage of the parametric bootstrap on privatized data ignored or avoided handling the effect of clamping, a technique employed by the majority of privacy mechanisms. Ignoring th...

Interpreting Differential Privacy in Terms of Disclosure Risk

As the use of differential privacy DP becomes widespread, the development of effective tools for reasoning about the privacy guarantee becomes increasingly critical. In pursuit of this goal, we demonstrate novel relationships between DP and measures of statistical disclosure risk. We suggest how...

Beyond the Worst Case: Extending Differential Privacy Guarantees to Realistic Adversaries

Differential Privacy DP is a family of definitions that bound the worst-case privacy leakage of a mechanism. One important feature of the worst-case DP guarantee is it naturally implies protections against adversaries with less prior information, more sophisticated attack goals, and complex...

AdeptHEQ-FL: Adaptive Homomorphic Encryption for Federated Learning of Hybrid Classical-Quantum Models with Dynamic Layer Sparing

Federated Learning FL faces inherent challenges in balancing model performance, privacy preservation, and communication efficiency, especially in non-IID decentralized environments. Recent approaches either sacrifice formal privacy guarantees, incur high overheads, or overlook quantum-enhanced...

Unifying Re-Identification, Attribute Inference, and Data Reconstruction Risks in Differential Privacy

Differentially private DP mechanisms are difficult to interpret and calibrate because existing methods for mapping standard privacy parameters to concrete privacy risks -- re-identification, attribute inference, and data reconstruction -- are both overly pessimistic and inconsistent. In this work...

Privacy-Utility-Fairness: a Balanced Approach to Vehicular-Traffic Management System

Location-based vehicular traffic management faces significant challenges in protecting sensitive geographical data while maintaining utility for traffic management and fairness across regions. Existing state-of-the-art solutions often fail to meet the required level of protection against linkage...

LDP$^3$: an Extensible and Multi-Threaded Toolkit for Local Differential Privacy Protocols and Post-Processing Methods

Local differential privacy LDP has become a prominent notion for privacy-preserving data collection. While numerous LDP protocols and post-processing PP methods have been developed, selecting an optimal combination under different privacy budgets and datasets remains a challenge. Moreover, the la...

Post-Processing in Local Differential Privacy: an Extensive Evaluation and Benchmark Platform

Local differential privacy LDP has recently gained prominence as a powerful paradigm for collecting and analyzing sensitive data from users' devices. However, the inherent perturbation added by LDP protocols reduces the utility of the collected data. To mitigate this issue, several post-processin...

Efficient Unlearning with Privacy Guarantees

Privacy protection laws, such as the GDPR, grant individuals the right to request the forgetting of their personal data not only from databases but also from machine learning ML models trained on them. Machine unlearning has emerged as a practical means to facilitate model forgetting of data...

The Landscape of Memorization in LLMs: Mechanisms, Measurement, and Mitigation

Large Language Models LLMs have demonstrated remarkable capabilities across a wide range of tasks, yet they also exhibit memorization of their training data. This phenomenon raises critical questions about model behavior, privacy risks, and the boundary between learning and memorization. Addressi...

UniAud: a Unified Auditing Framework for High Auditing Power and Utility with One Training Run

Differentially private DP optimization has been widely adopted as a standard approach to provide rigorous privacy guarantees for training datasets. DP auditing verifies whether a model trained with DP optimization satisfies its claimed privacy level by estimating empirical privacy lower bounds...

Model Inversion Attacks on Llama 3: Extracting PII from Large Language Models

Large language models LLMs have transformed natural language processing, but their ability to memorize training data poses significant privacy risks. This paper investigates model inversion attacks on the Llama 3.2 model, a multilingual LLM developed by Meta. By querying the model with carefully...