The Double-edged Sword of LLM-based Data Reconstruction: Understanding and Mitigating Contextual Vulnerability in Word-level Differential Privacy Text Sanitization

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Meisenbacher, Stephen, Klymenko, Alexandra, Bodea, Andreea-Elena, Matthes, Florian
Format:	Preprint
Published:	2025
Subjects:	Cryptography and Security Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2508.18976