Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation

• URL: http://arxiv.org/abs/2508.03098v1
• Date: Tue, 05 Aug 2025 05:22:13 GMT
• Title: Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation
• Authors: Haoran Wang, Xiongxiao Xu, Baixiang Huang, Kai Shu,
• Abstract summary: Privacy-Aware Decoding (PAD) is a lightweight, inference-time defense that adaptively injects calibrated Gaussian noise into token logits during generation.<n>PAD integrates confidence-based screening to selectively protect high-risk tokens, efficient sensitivity estimation to minimize unnecessary noise, and context-aware noise calibration to balance privacy with generation quality.<n>Our work takes an important step toward mitigating privacy risks in RAG via decoding strategies, paving the way for universal and scalable privacy solutions in sensitive domains.
• Score: 26.573578326262307
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Retrieval-Augmented Generation (RAG) enhances the factual accuracy of large language models (LLMs) by conditioning outputs on external knowledge sources. However, when retrieval involves private or sensitive data, RAG systems are susceptible to extraction attacks that can leak confidential information through generated responses. We propose Privacy-Aware Decoding (PAD), a lightweight, inference-time defense that adaptively injects calibrated Gaussian noise into token logits during generation. PAD integrates confidence-based screening to selectively protect high-risk tokens, efficient sensitivity estimation to minimize unnecessary noise, and context-aware noise calibration to balance privacy with generation quality. A \renyi Differential Privacy (RDP) accountant rigorously tracks cumulative privacy loss, enabling explicit per-response $(\varepsilon, \delta)$-DP guarantees for sensitive outputs. Unlike prior approaches requiring retraining or corpus-level filtering, PAD is model-agnostic and operates entirely at decoding time with minimal computational overhead. Experiments on three real-world datasets demonstrate that PAD substantially reduces private information leakage while preserving response utility, outperforming existing retrieval- and post-processing-based defenses. Our work takes an important step toward mitigating privacy risks in RAG via decoding strategies, paving the way for universal and scalable privacy solutions in sensitive domains. Our code is available: https://github.com/wang2226/PAD.

Related papers

• Improving Noise Efficiency in Privacy-preserving Dataset Distillation [59.57846442477106]
  We introduce a novel framework that decouples sampling from optimization for better convergence and improves signal quality.<n>On CIFAR-10, our method achieves a textbf10.0% improvement with 50 images per class and textbf8.3% increase with just textbfone-fifth the distilled set size of previous state-of-the-art methods.
  arXiv  Detail & Related papers  (2025-08-03T13:15:52Z)
• Privacy-Preserving Federated Embedding Learning for Localized Retrieval-Augmented Generation [60.81109086640437]
  We propose a novel framework called Federated Retrieval-Augmented Generation (FedE4RAG)<n>FedE4RAG facilitates collaborative training of client-side RAG retrieval models.<n>We apply homomorphic encryption within federated learning to safeguard model parameters.
  arXiv  Detail & Related papers  (2025-04-27T04:26:02Z)
• $(ε, δ)$-Differentially Private Partial Least Squares Regression [1.8666451604540077]
  We propose an $(epsilon, delta)$-differentially private PLS (edPLS) algorithm to ensure the privacy of the data underlying the model.<n> Experimental results demonstrate that edPLS effectively renders privacy attacks, aimed at recovering unique sources of variability in the training data.
  arXiv  Detail & Related papers  (2024-12-12T10:49:55Z)
• Privacy-Preserving Retrieval-Augmented Generation with Differential Privacy [25.896416088293908]
  retrieval-augmented generation (RAG) is particularly effective in assisting large language models (LLMs)<n>RAG outputs risk leaking sensitive information from the external data source.<n>We propose an algorithm that smartly spends privacy budget only for the tokens that require the sensitive information.
  arXiv  Detail & Related papers  (2024-12-06T01:20:16Z)
• Attack-Aware Noise Calibration for Differential Privacy [11.222654178949234]
  Differential privacy (DP) is a widely used approach for mitigating privacy risks when training machine learning models on sensitive data. The scale of the added noise is critical, as it determines the trade-off between privacy and utility. We show that first calibrating the noise scale to a privacy budget $varepsilon$, and then translating epsilon to attack risk leads to overly conservative risk assessments.
  arXiv  Detail & Related papers  (2024-07-02T11:49:59Z)
• Adaptive Differential Privacy in Federated Learning: A Priority-Based Approach [0.0]
  Federated learning (FL) develops global models without direct access to local datasets. DP offers a framework that gives a privacy guarantee by adding certain amounts of noise to parameters. We propose adaptive noise addition in FL which decides the value of injected noise based on features' relative importance.
  arXiv  Detail & Related papers  (2024-01-04T03:01:15Z)
• TeD-SPAD: Temporal Distinctiveness for Self-supervised Privacy-preservation for video Anomaly Detection [59.04634695294402]
  Video anomaly detection (VAD) without human monitoring is a complex computer vision task. Privacy leakage in VAD allows models to pick up and amplify unnecessary biases related to people's personal information. We propose TeD-SPAD, a privacy-aware video anomaly detection framework that destroys visual private information in a self-supervised manner.
  arXiv  Detail & Related papers  (2023-08-21T22:42:55Z)
• A Randomized Approach for Tight Privacy Accounting [63.67296945525791]
  We propose a new differential privacy paradigm called estimate-verify-release (EVR) EVR paradigm first estimates the privacy parameter of a mechanism, then verifies whether it meets this guarantee, and finally releases the query output. Our empirical evaluation shows the newly proposed EVR paradigm improves the utility-privacy tradeoff for privacy-preserving machine learning.
  arXiv  Detail & Related papers  (2023-04-17T00:38:01Z)
• Private Reinforcement Learning with PAC and Regret Guarantees [69.4202374491817]
  We design privacy preserving exploration policies for episodic reinforcement learning (RL) We first provide a meaningful privacy formulation using the notion of joint differential privacy (JDP) We then develop a private optimism-based learning algorithm that simultaneously achieves strong PAC and regret bounds, and enjoys a JDP guarantee.
  arXiv  Detail & Related papers  (2020-09-18T20:18:35Z)
• RDP-GAN: A R\'enyi-Differential Privacy based Generative Adversarial Network [75.81653258081435]
  Generative adversarial network (GAN) has attracted increasing attention recently owing to its impressive ability to generate realistic samples with high privacy protection. However, when GANs are applied on sensitive or private training examples, such as medical or financial records, it is still probable to divulge individuals' sensitive and private information. We propose a R'enyi-differentially private-GAN (RDP-GAN), which achieves differential privacy (DP) in a GAN by carefully adding random noises on the value of the loss function during training.
  arXiv  Detail & Related papers  (2020-07-04T09:51:02Z)
• Mitigating Query-Flooding Parameter Duplication Attack on Regression Models with High-Dimensional Gaussian Mechanism [12.017509695576377]
  Differential privacy (DP) has been considered a promising technique to mitigate this attack. We show that the adversary can launch a query-flooding parameter duplication (QPD) attack to infer the model information. We propose a novel High-Dimensional Gaussian (HDG) mechanism to prevent unauthorized information disclosure.
  arXiv  Detail & Related papers  (2020-02-06T01:47:08Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.