Research map / Learning theory and algorithms / Fairness
Decoupling Bias, Aligning Distributions: Synergistic Fairness Optimization for Deepfake Detection
Feng Ding, Wenhui Yi, Yunpeng Zhou, Xinan He, Hong Rao, Shu Hu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Global Distribution Alignment to the baseline (ablation) | Xception · FF++ test | AUC (%): 92.69 -> 97.22 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · FF++ test | AUC (%): 97.22 -> 97.71 | Improved |
| add Global Distribution Alignment to the baseline (ablation) | Xception · DFDC | AUC (%): 56.13 -> 60.91 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · DFDC | AUC (%): 60.91 -> 61.86 | Improved |
| add Global Distribution Alignment to the baseline (ablation) | Xception · Celeb-DF | AUC (%): 69.18 -> 72.89 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · Celeb-DF | AUC (%): 72.89 -> 76.75 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · DFD | AUC (%): 77.78 -> 81.46 | Improved |
| add Global Distribution Alignment to the baseline (ablation) | ResNet-50 · DFD | AUC (%): 71.31 -> 70.99 | Hurt |
| add Structural Fairness Decoupling on top of GDA (ablation) | ResNet-50 · DFD | AUC (%): 70.99 -> 78.63 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · FF++ test, gender | F_FPR (%): 3.91 -> 0.53 | Improved |
| add Structural Fairness Decoupling on top of GDA (ablation) | Xception · FF++ test, race | F_FPR (%): 3.95 -> 9.29 | Hurt |
| add Global Distribution Alignment to the baseline (ablation) | ResNet-50 · FF++ test, gender | F_DP (%): 9.27 -> 5.54 | Improved |
Closest papers by meaning
- Fair Deepfake Detectors Can Generalize (NeurIPS 2025 · similarity 0.91)
- Preserving Fairness Generalization in Deepfake Detection (CVPR 2024 · similarity 0.90)
- Towards Harmless Rawlsian Fairness Regardless of Demographic Prior (NeurIPS 2024 · similarity 0.87)
- Fairness-aware Anomaly Detection via Fair Projection (NeurIPS 2025 · similarity 0.86)
- On the Fairness ROAD: Robust Optimization for Adversarial Debiasing (ICLR 2024 · similarity 0.85)
- Exploring Unbiased Deepfake Detection via Token-Level Shuffling and Mixing (AAAI 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
