PSY: Posterior Sampling Based Privacy Enhancer in Large Language Models

Privacy vulnerabilities in LLMs, such as leakage from memorization, have been constantly identified, and various mitigation proposals have been proposed. LoRA is usually used in fine-tuning LLMs and a good entry point to insert privacy-enhancing modules. In this ongoing research, we introduce PSY, a...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2024-10
Hauptverfasser:	Sun, Yulian, Li, Duan, Li, Yong
Format:	Artikel
Sprache:	eng
Schlagworte:	Large language models Leakage Privacy Sampling
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!