Research map / Continual learning / Online and replay-based
Generalization-Preserved Learning: Closing the Backdoor to Catastrophic Forgetting in Continual Deepfake Detection
Xueyi Zhang, Peiyin Zhu, Chengwei Zhang, Zhiyuan Yan, Jikang Cheng, Mingrui Lao, Siqi Cai, Yanming Guo
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 Generalized Gradient Projection (ablation) | GPL · Continual deepfake detection S1 | AF (%): 11.79 -> 1.42 | Improved |
| add Hyperbolic Visual Alignment (ablation) | GPL · Continual deepfake detection S1 | AA (%): 82.80 -> 92.14 | Improved |
| add GGP and HVA to watermark-generator baseline (ablation) | watermark generator · Continual deepfake detection S1 | AA (%): 78.88 -> 92.14 | Improved |
| add replay data (ablation) | GPL · Continual deepfake detection S1 | AA (%): 92.14 -> 92.16 | No clear effect |
| use hyperbolic instead of Euclidean space (ablation) | GPL · Continual deepfake detection S1 | AA (%): 88.80 -> 92.14 | Improved |
| use hyperbolic instead of Euclidean space (lower forgetting) (ablation) | GPL · Continual deepfake detection S1 | AF (%): 2.87 -> 1.42 | Improved |
| use GPL instead of DFIL | DFIL · Unseen datasets | Avg. ACC (%): 64.46 -> 82.84 | Improved |
| add GGP (ablation) | GPL · Unseen datasets | Avg. ACC (%): 76.82 -> 82.84 | Improved |
| add HVA (ablation) | GPL · Unseen datasets | Avg. ACC (%): 74.48 -> 82.84 | Improved |
| use GPL instead of DFIL | DFIL · Training order S2 | AA (%): 87.10 -> 89.94 | Improved |
| use GPL instead of DFIL | DFIL · Training order S3 | AF (%): 1.70 -> 1.76 | No clear effect |
Closest papers by meaning
- Veritas: Generalizable Deepfake Detection via Pattern-Aware Reasoning (ICLR 2026 · similarity 0.89)
- Transcending Forgery Specificity with Latent Space Augmentation for Generalizable Deepfake Detection (CVPR 2024 · similarity 0.88)
- Generalizing Deepfake Video Detection with Plug-and-Play: Video-Level Blending and Spatiotemporal Adapter Tuning (CVPR 2025 · similarity 0.87)
- DiffusionFake: Enhancing Generalization in Deepfake Detection via Guided Stable Diffusion (NeurIPS 2024 · similarity 0.87)
- Towards More General Video-based Deepfake Detection through Facial Component Guided Adaptation for Foundation Model (CVPR 2025 · similarity 0.87)
- Exploring Unbiased Deepfake Detection via Token-Level Shuffling and Mixing (AAAI 2025 · similarity 0.87)
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.
