PANOPTICON: A PII-Based Assemblage of Naturalistic Output Tokens for Investigating Privacy Leakage Within LLM Context Window
Ryan Thornton, Mir Mehedi Ahsan Pritom, Maanak Gupta
cs.AI, cs.CR
Submitted: 2026-07-17
Comments: 9 pages, 3 figures, 3 tables. Submitted to IEEE CARS 2026
License: http://creativecommons.org/licenses/by/4.0/
The gist: Large Language Models (LLMs) are capable of generalizing human language for the completion of never-before-seen tasks, leading to widespread deployment.
Terminology
Abstract
Large Language Models (LLMs) are capable of generalizing human language for the completion of never-before-seen tasks, leading to widespread deployment. While this automation provides clear utility, completing these tasks often requires the insertion of Personally Identifiable Information (PII), strings of information that uniquely identify some individual, raising privacy concerns. However, ethics has prevented the curation of a public, authentic dataset of PII. Without an appropriate dataset, it is difficult to quantify privacy risks. Thus, we introduce the PANOPTICON pipeline and dataset. The dataset, generated by Meta's Llama-3.1-8B-Instruct model, contains 67, 718 prompts, intended for the models context window, containing PII spans derived from 9,674 publicly available synthetic user profiles. We measure lexical diversity and S-BERT diversity of the created dataset to evaluate realism. Finally, we present a case study showcasing the utility of PANOPTICON data for understanding Prompt Inversion Attacks (PIAs). PANOPTICON thus emerges as the first benchmark dataset for studying PIAs over private corpora, providing a foundation for future LLM privacy research.
Sources
- A Survey of Large Language Models
- Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG
- A Survey on Large Language Models for Critical Societal Domains: Finance, Healthcare, and Law
- Position: Privacy Is Not Just Memorization!
- PANORAMA: A synthetic PII-laced dataset for studying sensitive data memorization in LLMs
- DORY: Deliberative Prompt Recovery for LLM
- Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theory
- The Pile: An 800GB Dataset of Diverse Text for Language Modeling
- Quantifying Memorization Across Neural Language Models
- Differentially Private Decoding in Large Language Models
- Dual-Priv Pruning : Efficient Differential Private Fine-Tuning in Multimodal Large Language Models
- An Evaluation Study of Hybrid Methods for Multilingual PII Detection
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
- VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection