DP-MGTD: Privacy-Preserving Machine-Generated Text Detection via Adaptive Differentially Private Entity Sanitization

TL;DR

DP-MGTD introduces an adaptive differential privacy framework for detecting machine-generated text that balances privacy with high detection accuracy, leveraging noise mechanisms to enhance distinguishability.

Contribution

The paper presents a novel adaptive DP-based sanitization method that improves MGT detection accuracy while preserving privacy, addressing limitations of existing anonymization and DP techniques.

Findings

Achieves near-perfect detection accuracy under strict privacy guarantees.

Outperforms non-private baselines significantly.

Utilizes a two-stage noise calibration mechanism.

Abstract

The deployment of Machine-Generated Text (MGT) detection systems necessitates processing sensitive user data, creating a fundamental conflict between authorship verification and privacy preservation. Standard anonymization techniques often disrupt linguistic fluency, while rigorous Differential Privacy (DP) mechanisms typically degrade the statistical signals required for accurate detection. To resolve this dilemma, we propose \textbf{DP-MGTD}, a framework incorporating an Adaptive Differentially Private Entity Sanitization algorithm. Our approach utilizes a two-stage mechanism that performs noisy frequency estimation and dynamically calibrates privacy budgets, applying Laplace and Exponential mechanisms to numerical and textual entities respectively. Crucially, we identify a counter-intuitive phenomenon where the application of DP noise amplifies the distinguishability between human and machine text by exposing distinct sensitivity patterns to perturbation. Extensive experiments on the MGTBench-2.0 dataset show that our method achieves near-perfect detection accuracy, significantly outperforming non-private baselines while satisfying strict privacy guarantees.