通过重加权模型输出层输入，提升深度伪造检测的公平性。
Fair-FLIP: Fair Deepfake Detection with Fairness-Oriented Final Layer Input Prioritising
- 在训练后通过优先处理低变异性输入，降低不同群体间的检测偏差。
- 公平性指标最高提升30%，准确率仅下降0.25%。
- 适合关注检测系统公平性的研究者与应用开发者。
人工智能生成内容日益普及，但其恶意使用（尤其是深度伪造）严重威胁公众信任与公共讨论。尽管现有深度伪造检测方法预测性能优异，但在种族、性别等人口属性上常表现出偏差。本文提出一种新型后处理方法——面向公平性的最终层输入优先策略（Fair-FLIP），通过重新加权已训练模型的最终层输入，减少子群体间的差异，优先保留低变异性输入，抑制高变异性输入。实验表明，相比基线（无公平性优化）和当前最优方法，Fair-FLIP可将公平性指标提升最高达30%，同时保持基线准确率，仅带来0.25%的微小下降。代码已开源：https://github.com/szandala/fair-deepfake-detection-toolbox
原文摘要 · Abstract (English)
Artificial Intelligence-generated content has become increasingly popular, yet its malicious use, particularly the deepfakes, poses a serious threat to public trust and discourse. While deepfake detection methods achieve high predictive performance, they often exhibit biases across demographic attributes such as ethnicity and gender. In this work, we tackle the challenge of fair deepfake detection, aiming to mitigate these biases while maintaining robust detection capabilities. To this end, we propose a novel post-processing approach, referred to as Fairness-Oriented Final Layer Input Prioritising (Fair-FLIP), that reweights a trained model's final-layer inputs to reduce subgroup disparities, prioritising those with low variability while demoting highly variable ones. Experimental results comparing Fair-FLIP to both the baseline (without fairness-oriented de-biasing) and state-of-the-art approaches show that Fair-FLIP can enhance fairness metrics by up to 30% while maintaining baseline accuracy, with only a negligible reduction of 0.25%. Code is available on Github: https://github.com/szandala/fair-deepfake-detection-toolbox
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。