English

Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation

Cryptography and Security 2026-04-14 v2 Artificial Intelligence

Abstract

Large language models (LLMs) have emerged as a powerful tool for synthetic data generation. A particularly important use case is producing synthetic replicas of private text, which requires carefully balancing privacy and utility. We propose Realistic and Privacy-Preserving Synthetic Data Generation (RPSG), which uses private seeds and integrates privacy-preserving strategies, including a formal differential privacy (DP) mechanism in the candidate selection, to generate realistic synthetic data. Comprehensive experiments against state-of-the-art private synthetic data generation methods demonstrate that RPSG achieves high fidelity to private data while providing strong privacy protection.

Keywords

Cite

@article{arxiv.2604.07486,
  title  = {Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation},
  author = {Qian Ma and Sarah Rajtmajer},
  journal= {arXiv preprint arXiv:2604.07486},
  year   = {2026}
}

Comments

22 pages, 7 figures, 18 tables