Prism: Private Relational Data Synthesis with Language Models
Proceedings of the ACM on Management of Data · 18.05.2026 · 4 (3) · 1-28
DOI: 10.1145/3802102 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the ACM on Management of Data, online erschienen am 18.05.2026. Verfasst von Guohui Guan und 1 weiteren Autor:innen. Die Arbeit nennt 103 Literaturangaben.
Abstract (Originalsprache)
Organizations increasingly rely on relational data containing sensitive personal information for downstream tasks. A common approach is to release synthetic data as a privacy-preserving substitute, typically by learning a generative model and sampling from it. Differential privacy (DP) provides a formal guarantee of data privacy, and existing data synthesis systems aim to balance its inherent trade-off between data privacy and task utility. The advent of pretrained large language models (LLMs) has reshaped this trade-off landscape: on the utility side, LLMs possess stronger representational capabilities and encode extensive prior knowledge that does not consume the privacy budget, leading to improved task utility; on the other hand, enforcing differential privacy for LLM-based relational data synthesis introduces new technical challenges. We present P rism, an end-to-end framework for DP relational data synthesis using LLMs. P rism privatizes knowledge transfer from an ensemble of teacher LLMs to a student generator via DP output aggregation. We introduce three key innovations in the aggregation mechanism to address LLM-specific challenges, including 1) token pruning with tries to minimize teacher queries and hence the privacy cost; 2) a data-dependent DP mechanism that adaptively scales noise for efficient budget use; and 3) probabilistic sampling with threshold filtering to reduce bias and preserve diversity. Extensive empirical results across twelve experiments show that Prism achieves substantially higher predictive utility than eight state-of-the-art methods under the same privacy budget.
Zitieren
APA
Guan, G., & Ge, C. (2026). Prism: Private Relational Data Synthesis with Language Models. Proceedings of the ACM on Management of Data, 4(3), 1-28. https://doi.org/10.1145/3802102
Vancouver
Guan G, Ge C. Prism: Private Relational Data Synthesis with Language Models. Proceedings of the ACM on Management of Data. 2026;4(3):1-28. doi:10.1145/3802102
BibTeX
@article{Guan2026,
author = {Guan, Guohui and Ge, Chang},
title = {Prism: Private Relational Data Synthesis with Language Models},
journal = {Proceedings of the ACM on Management of Data},
year = {2026},
volume = {4},
number = {3},
pages = {1--28},
doi = {10.1145/3802102}
}
EUVIMED