Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch

• URL: http://arxiv.org/abs/2602.03183v1
• Date: Tue, 03 Feb 2026 06:54:46 GMT
• Title: Privasis: Synthesizing the Largest "Public" Private Dataset from Scratch
• Authors: Hyunwoo Kim, Niloofar Mireshghallah, Michael Duan, Rui Xin, Shuyue Stella Li, Jaehun Jung, David Acuna, Qi Pang, Hanshen Xiao, G. Edward Suh, Sewoong Oh, Yulia Tsvetkov, Pang Wei Koh, Yejin Choi,
• Abstract summary: We present Privasis, the first million-scale fully synthetic dataset entirely built from scratch.<n>Compared to existing datasets, Privasis offers orders-of-magnitude larger scale with quality.<n>We leverage Privasis to construct a parallel corpus for text sanitization with our pipeline that decomposes texts and applies targeted sanitization.
• Score: 101.49955223689268
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Research involving privacy-sensitive data has always been constrained by data scarcity, standing in sharp contrast to other areas that have benefited from data scaling. This challenge is becoming increasingly urgent as modern AI agents--such as OpenClaw and Gemini Agent--are granted persistent access to highly sensitive personal information. To tackle this longstanding bottleneck and the rising risks, we present Privasis (i.e., privacy oasis), the first million-scale fully synthetic dataset entirely built from scratch--an expansive reservoir of texts with rich and diverse private information--designed to broaden and accelerate research in areas where processing sensitive social data is inevitable. Compared to existing datasets, Privasis, comprising 1.4 million records, offers orders-of-magnitude larger scale with quality, and far greater diversity across various document types, including medical history, legal documents, financial records, calendars, and text messages with a total of 55.1 million annotated attributes such as ethnicity, date of birth, workplace, etc. We leverage Privasis to construct a parallel corpus for text sanitization with our pipeline that decomposes texts and applies targeted sanitization. Our compact sanitization models (<=4B) trained on this dataset outperform state-of-the-art large language models, such as GPT-5 and Qwen-3 235B. We plan to release data, models, and code to accelerate future research on privacy-sensitive domains and agents.

Related papers

• How to DP-fy Your Data: A Practical Guide to Generating Synthetic Data With Differential Privacy [52.00934156883483]
  Differential Privacy (DP) is a framework for reasoning about and limiting information leakage.<n>Differentially Private Synthetic data refers to synthetic data that preserves the overall trends of source data.
  arXiv  Detail & Related papers  (2025-12-02T21:14:39Z)
• MAGPIE: A dataset for Multi-AGent contextual PrIvacy Evaluation [54.410825977390274]
  Existing benchmarks to evaluate contextual privacy in LLM-agents primarily assess single-turn, low-complexity tasks.<n>We first present a benchmark - MAGPIE comprising 158 real-life high-stakes scenarios across 15 domains.<n>We then evaluate the current state-of-the-art LLMs on their understanding of contextually private data and their ability to collaborate without violating user privacy.
  arXiv  Detail & Related papers  (2025-06-25T18:04:25Z)
• PrivacyLens: Evaluating Privacy Norm Awareness of Language Models in Action [54.11479432110771]
  PrivacyLens is a novel framework designed to extend privacy-sensitive seeds into expressive vignettes and further into agent trajectories.<n>We instantiate PrivacyLens with a collection of privacy norms grounded in privacy literature and crowdsourced seeds.<n>State-of-the-art LMs, like GPT-4 and Llama-3-70B, leak sensitive information in 25.68% and 38.69% of cases, even when prompted with privacy-enhancing instructions.
  arXiv  Detail & Related papers  (2024-08-29T17:58:38Z)
• NAP^2: A Benchmark for Naturalness and Privacy-Preserving Text Rewriting by Learning from Human [56.46355425175232]
  We suggest sanitizing sensitive text using two common strategies used by humans.<n>We curate the first corpus, coined NAP2, through both crowdsourcing and the use of large language models.<n>Compared to the prior works on anonymization, the human-inspired approaches result in more natural rewrites.
  arXiv  Detail & Related papers  (2024-06-06T05:07:44Z)
• Practical considerations on using private sampling for synthetic data [1.3654846342364308]
  Differential privacy for synthetic data generation has received much attention due to the ability of preserving privacy while freely using the synthetic data. Private sampling is the first noise-free method to construct differentially private synthetic data with rigorous bounds for privacy and accuracy. We provide an implementation of the private sampling algorithm and discuss the realism of its constraints in practical cases.
  arXiv  Detail & Related papers  (2023-12-12T10:20:04Z)
• A Unified View of Differentially Private Deep Generative Modeling [60.72161965018005]
  Data with privacy concerns comes with stringent regulations that frequently prohibited data access and data sharing. Overcoming these obstacles is key for technological progress in many real-world application scenarios that involve privacy sensitive data. Differentially private (DP) data publishing provides a compelling solution, where only a sanitized form of the data is publicly released.
  arXiv  Detail & Related papers  (2023-09-27T14:38:16Z)
• More Data Types More Problems: A Temporal Analysis of Complexity, Stability, and Sensitivity in Privacy Policies [0.0]
  Data brokers and data processors are part of a multi-billion-dollar industry that profits from collecting, buying, and selling consumer data. Yet there is little transparency in the data collection industry which makes it difficult to understand what types of data are being collected, used, and sold.
  arXiv  Detail & Related papers  (2023-02-17T15:21:24Z)
• Synthetic Text Generation with Differential Privacy: A Simple and Practical Recipe [32.63295550058343]
  We show that a simple and practical recipe in the text domain is effective in generating useful synthetic text with strong privacy protection. Our method produces synthetic text that is competitive in terms of utility with its non-private counterpart.
  arXiv  Detail & Related papers  (2022-10-25T21:21:17Z)
• You Are What You Write: Preserving Privacy in the Era of Large Language Models [2.3431670397288005]
  We present an empirical investigation into the extent of the personal information encoded into pre-trained representations by a range of popular models. We show a positive correlation between the complexity of a model, the amount of data used in pre-training, and data leakage.
  arXiv  Detail & Related papers  (2022-04-20T11:12:53Z)

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.