Decoupling Bias, Aligning Distributions: Synergistic Fairness Optimization for Deepfake Detection

TL;DR

This paper introduces a dual-mechanism framework that enhances fairness in deepfake detection models by decoupling demographic biases and aligning distributions, achieving better fairness without sacrificing accuracy.

Contribution

It proposes a novel collaborative optimization method combining structural fairness decoupling and global distribution alignment for fairer deepfake detection.

Findings

Improves inter-group and intra-group fairness

Maintains detection accuracy across domains

Outperforms existing fairness methods

Abstract

Fairness is a core element in the trustworthy deployment of deepfake detection models, especially in the field of digital identity security. Biases in detection models toward different demographic groups, such as gender and race, may lead to systemic misjudgments, exacerbating the digital divide and social inequities. However, current fairness-enhanced detectors often improve fairness at the cost of detection accuracy. To address this challenge, we propose a dual-mechanism collaborative optimization framework. Our proposed method innovatively integrates structural fairness decoupling and global distribution alignment: decoupling channels sensitive to demographic groups at the model architectural level, and subsequently reducing the distance between the overall sample distribution and the distributions corresponding to each demographic group at the feature level. Experimental results demonstrate that, compared with other methods, our framework improves both inter-group and intra-group fairness while maintaining overall detection accuracy across domains. The code is available at https://github.com/ywh1093/Fairness-Optimization.