mlsanityData to 7 Oct 2026

Is External Database Protection Static in Retrieval-Augmented Generation? Rethinking Privacy Preservation under Dynamic Queries

arXiv 2607.14811 ↗First posted 16 Jul 2026cs.CRcited by 0

On the map: Retrieval-augmented generation → Privacy and security of RAG

Abstract (arXiv)

Retrieval-augmented generation (RAG) enhances large language models via external document retrieval, but retrieved contexts may leak sensitive information. Current privacy protection methods typically rely on a document-level static risk assumption, treating all retrieved documents as having the same privacy leakage risk. However, this assumption overlooks a fundamental characteristic of RAG: the privacy risk of a document is highly dependent on the user's query, making privacy leakage inherently query-driven and dynamic. To address this challenge, we propose a Prompt-Aware Dynamic Hierarchical Differential Privacy framework (PA-HDP) for privacy-preserving RAG. PA-HDP first performs a prompt-aware risk hierarchy to dynamically assess privacy risks under different queries. It then applies adaptive sensitive entity replacement and exponential mechanism-based text selection to provide differentiated privacy protection while preserving semantic utility. By protecting only the content that is truly sensitive under a given query, PA-HDP minimizes unnecessary modifications to the retrieval corpus. Extensive experiments on benchmark datasets demonstrate that PA-HDP significantly reduces privacy leakage while maintaining high retrieval quality, achieving a better privacy-utility trade-off than prior methods.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Privacy and security of RAG, one of 15 directions in Retrieval-augmented generation. Builds on 22 of its 41 references that are in the map. First posted on arXiv 16 Jul 2026.

Builds on

22 of 41 references.

Closest ancestors · 15

Same direction, newest first

  1. 2026Towards Secure Retrieval-Augmented Generation: A Comprehensive Review of Threats, Defenses and Benchmarks23 Mar
  2. 2026Differentially Private Retrieval-Augmented Generation16 Feb
  3. 2026Efficient Privacy-Preserving Retrieval Augmented Generation with Distance-Preserving Encryption18 Jan
  4. 2025Privacy Challenges and Solutions in Retrieval-Augmented Generation-Enhanced LLMs for Healthcare Chatbots: A Review of Applications, Risks, and Future Directions14 Nov
  5. 2025Private-RAG: Answering Multiple Queries with LLMs while Keeping Your Data Private10 Nov
  6. 2025Differentially Private Synthetic Text Generation for Retrieval-Augmented Generation (RAG)8 Oct
  7. 2025Privacy-Preserving Reasoning with Knowledge-Distilled Parametric Retrieval Augmented Generation1 Sep
  8. 2025Privacy-Aware Decoding: Mitigating Privacy Leakage of Large Language Models in Retrieval-Augmented Generation5 Aug
  9. 2025InvisibleInk: High-Utility and Low-Cost Text Generation with Differential Privacy30 Jun
  10. 2025Learning to Erase Private Knowledge from Multi-Documents for Retrieval-Augmented Large Language Models14 Apr
  11. 2024RAG with Differential Privacy26 Dec
  12. 2024RemoteRAG: A Privacy-Preserving LLM Cloud RAG Service17 Dec
3 more
  1. 2024Privacy-Preserving Retrieval-Augmented Generation with Differential Privacy6 Dec
  2. 2024Mitigating the Privacy Issues in Retrieval-Augmented Generation (RAG) via Pure Synthetic Data20 Jun
  3. 2024The Good and The Bad: Exploring Privacy Issues in Retrieval-Augmented Generation (RAG)23 Feb

Other references · 7

  1. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  2. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,185
  3. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,582
  4. 2023Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Biascited by 230
  5. 2022ZeroGen: Efficient Zero-shot Learning via Dataset Generationcited by 209
  6. 2023Privacy Implications of Retrieval-Based Language Modelscited by 31
  7. 2023Have it your way: Individualized Privacy Assignment for DP-SGDcited by 21

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Not All Entities are Created Equal: A Dynamic Anonymization Framework for Privacy-Preserving RAG
  2. 2026Security and Privacy in Retrieval-Augmented Generation: Architectures, Threats, Defenses, and Future Directions for Building Trustworthy Systems
  3. 2026P2RAG: Efficient Privacy-Preserving RAG Service Supporting Arbitrary Top-k Retrieval
  4. 2025Privacy-Aware RAG: Secure and Isolated Knowledge Retrieval
  5. 2026PRAG: End-to-End Privacy-Preserving Retrieval-Augmented Generation
  6. 2026Privacy-Preserving RAG by Concealing Sensitive Information from External LLMs
  7. 2026SoK: Privacy Risks and Mitigations in Retrieval-Augmented Generation Systems
  8. 2026SD-RAG: A Prompt-Injection-Resilient Framework for Selective Disclosure in Retrieval-Augmented Generation