mlsanityData to 7 Oct 2026

Hide and Seek (HaS): A Lightweight Framework for Prompt Privacy Protection

arXiv 2309.03057 ↗First posted 6 Sep 2023cs.CR cs.AIcited by 37

Not on the map. Why?

Abstract (arXiv)

Numerous companies have started offering services based on large language models (LLM), such as ChatGPT, which inevitably raises privacy concerns as users' prompts are exposed to the model provider. Previous research on secure reasoning using multi-party computation (MPC) has proven to be impractical for LLM applications due to its time-consuming and communication-intensive nature. While lightweight anonymization techniques can protect private information in prompts through substitution or masking, they fail to recover sensitive data replaced in the LLM-generated results. In this paper, we expand the application scenarios of anonymization techniques by training a small local model to de-anonymize the LLM's returned results with minimal computational overhead. We introduce the HaS framework, where "H(ide)" and "S(eek)" represent its two core processes: hiding private entities for anonymization and seeking private entities for de-anonymization, respectively. To quantitatively assess HaS's privacy protection performance, we propose both black-box and white-box adversarial models. Furthermore, we conduct experiments to evaluate HaS's usability in translation and classification tasks. The experimental findings demonstrate that the HaS framework achieves an optimal balance between privacy protection and utility.

Key facts. Cited by 37 papers in the map (data through 7 Oct 2026). Builds on 3 of its 9 references that are in the map. First posted on arXiv 6 Sep 2023.

Builds on

3 of 9 references.

References · 3

  1. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,592
  2. 2023QLoRA: Efficient Finetuning of Quantized LLMscited by 3,355
  3. 2019CTRL: A Conditional Transformer Language Model for Controllable Generationcited by 754

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2023: 2; 2024: 11; 2025: 13; 2026: 11.

37 papers.

Sort:

By area:

  1. 2023Split-and-Denoise: Protect large language model inference with local differential privacycited by 30
  2. 2024PrivacyRestore: Privacy-Preserving Inference in Large Language Models via Privacy Removal and Restorationcited by 27
  3. 2024Privacy Preserving Prompt Engineering: A Surveycited by 24
  4. 2024RemoteRAG: A Privacy-Preserving LLM Cloud RAG Servicecited by 17
  5. 2024Privacy in Fine-tuning Large Language Models: Attacks, Defenses, and Future Directionscited by 14
  6. 2024The Fire Thief Is Also the Keeper: Balancing Usability and Privacy in Promptscited by 13
  7. 2024EmojiPrompt: Generative Prompt Obfuscation for Privacy-Preserving Communication with Cloud-based LLMscited by 9
  8. 2024Casper: Prompt Sanitization for Protecting User Privacy in Web-Based Large Language Modelscited by 8
  9. 2025LLM Access Shield: Domain-Specific LLM Framework for Privacy Policy Compliancecited by 6
  10. 2025Towards Harnessing the Collaborative Power of Large and Small Models for Domain Taskscited by 4
  11. 2023ConfusionPrompt: Practical Private Inference for Online Large Language Modelscited by 3
  12. 2024PFID: Privacy First Inference Delegation Framework for LLMscited by 3
25 more
  1. 2025A General Pseudonymization Framework for Cloud-Based LLMs: Replacing Privacy Information in Controlled Text Generationcited by 3
  2. 2025MaskSQL: Safeguarding Privacy for LLM-Based Text-to-SQL via Abstractioncited by 3
  3. 2026Need to Know: Contextual-Integrity-Grounded Query Rewriting for Privacy-Conscious LLM Delegationcited by 3
  4. 2025LegalGuardian: A Privacy-Preserving Framework for Secure Integration of Large Language Models in Legal Practicecited by 2
  5. 2025DP-GTR: Differentially Private Prompt Protection via Group Text Rewritingcited by 2
  6. 2025Anti-adversarial Learning: Desensitizing Prompts for Large Language Modelscited by 2
  7. 2024Confidential Prompting: Privacy-preserving LLM Inference on Cloudcited by 1
  8. 2025Privacy Meets Explainability: Managing Confidential Data and Transparency Policies in LLM-Empowered Sciencecited by 1
  9. 2025Semantically-Aware LLM Agent to Enhance Privacy in Conversational AI Servicescited by 1
  10. 2026Your Inference Request Will Become a Black Box: Confidential Inference for Cloud-based Large Language Modelscited by 1
  11. 2026Towards Privacy-Preserving Machine Translation at the Inference Stage: A New Task and Benchmarkcited by 1
  12. 2026PAAC: Privacy-Aware Agentic Device-Cloud Collaborationcited by 1
  13. 2026PrivScope: Task-scoped Disclosure Control for Hybrid Agentic Systemscited by 1
  14. 2026Demystifying the Privacy-Utility Trade-off in LLM Interactionscited by 1
  15. 2024Navigating the Future of Federated Recommendation Systems with Foundation Modelscited by 0
  16. 2024Preempting Text Sanitization Utility in Resource-Constrained Privacy-Preserving LLM Interactionscited by 0
  17. 2025DP-GPL: Differentially Private Graph Prompt Learningcited by 0
  18. 2025Collaborative LLM Numerical Reasoning with Local Data Protectioncited by 0
  19. 2025Stereo-LiDAR Fusion by Semi-Global Matching With Discrete Disparity-Matching Cost and Semidensificationcited by 0
  20. 2025Semantic Encryption: Secure and Effective Interaction with Cloud-based Large Language Models via Semantic Transformationcited by 0
  21. 2026Differentially Private and Communication Efficient Large Language Model Split Inference via Stochastic Quantization and Soft Promptcited by 0
  22. 2026PromptGraph: Graph-Guided Prompt Sanitization for Balancing Privacy and Utility in LLM Inferencecited by 0
  23. 2026GuardianAgent: Policy-Conditioned Risk-Adaptive Anonymization with Verified Adversarial Escalationcited by 0
  24. 2026LLMs as Linguistic Chameleons: Decoupling Semantics and Structure for Privacy-Preserving Communicationcited by 0
  25. 2026Learning to Refer: Client-Resolved Generation for Privacy-Aware Language Modelscited by 0