Evaluating LLM Simulators as Differentially Private Data Generators
arXiv:2604.15461v1 Announce Type: new
Abstract: LLM-based simulators offer a promising path for generating complex synthetic data where traditional differentially private (DP) methods struggle with high-dimensional user profiles. But can LLMs faithful…