Does Prompt-Tuning Language Model Ensure Privacy?

by   Shangyu Xie, et al.

Prompt-tuning has received attention as an efficient tuning method in the language domain, i.e., tuning a prompt that is a few tokens long, while keeping the large language model frozen, yet achieving comparable performance with conventional fine-tuning. Considering the emerging privacy concerns with language models, we initiate the study of privacy leakage in the setting of prompt-tuning. We first describe a real-world email service pipeline to provide customized output for various users via prompt-tuning. Then we propose a novel privacy attack framework to infer users' private information by exploiting the prompt module with user-specific signals. We conduct a comprehensive privacy evaluation on the target pipeline to demonstrate the potential leakage from prompt-tuning. The results also demonstrate the effectiveness of the proposed attack.


Prefix-Tuning: Optimizing Continuous Prompts for Generation

Fine-tuning is the de facto way to leverage large pretrained language mo...

Domain Private Transformers

Large, general purpose language models have demonstrated impressive perf...

Making Split Learning Resilient to Label Leakage by Potential Energy Loss

As a practical privacy-preserving learning method, split learning has dr...

KART: Privacy Leakage Framework of Language Models Pre-trained with Clinical Records

Nowadays, mainstream natural language pro-cessing (NLP) is empowered by ...

Just Fine-tune Twice: Selective Differential Privacy for Large Language Models

With the increasing adoption of NLP models in real-world products, it be...

Rethinking Masked Language Modeling for Chinese Spelling Correction

In this paper, we study Chinese Spelling Correction (CSC) as a joint dec...

Learning to Customize Language Model for Generation-based dialog systems

Personalized conversation systems have received increasing attention rec...

Please sign up or login with your details

Forgot password? Click here to reset