AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

DFD-HR: Generalizable Deepfake Detection via Hierarchical Routing Learning

Jiamu Sun, Zhiyuan Yan, Ke-Yue Zhang, Taiping Yao, Shouhong Ding

CVPR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning

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 changedWhereMeasuredEffect
add Early Layer Pruning (ablation)DFD-HR (CLIP ViT-L/14) · CDF-v2/FaceShifter avgvideo AUC (%): 90.3 -> 91.3Improved
add Token Selection with Spearman rank loss (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 91.3 -> 92.4Improved
add token-level Expert Routing (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 92.4 -> 92.9Improved
add Multi-Scale Fusion (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 92.9 -> 93.6Improved
replace Token Selection with Sparse Attention (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 91.3 -> 91.0 (vs ELP baseline)Hurt
replace Token Selection with PCA or FastV (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 91.7 and 91.8 vs 92.4 for oursMixed
replace Multi-Scale Fusion with RINE or Multi CLS Pooling or D3 (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): baseline 90.3 -> 84.2 (RINE), 89.0, 89.1Hurt
remove Gumbel-Sigmoid hard routing in Early Layer Pruning (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 93.6 -> 74.8Hurt
share layer judge across layers (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 93.6 -> 91.8Hurt
use noisy gate instead of Top2 routing or balance loss in experts (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 92.9 (Top2), 92.4 (Balance) -> 93.6Improved
increase number of experts beyond 4 (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 93.6 (4) -> 93.1 (6), 93.0 (8)Hurt
increase rank loss weight lambda1 from 0.1 to 1.0 (ablation)DFD-HR · CDF-v2/FaceShifter avgvideo AUC (%): 93.6 -> 92.5Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy