AI Paper Atlas

Research map / Architectures and efficiency / Pruning and efficient tuning

SRRA: Stable-Rank-Based Residual Adaptation for Generalizable Deepfake Detection

Huakun Liu, Wenjie Li, Changsheng Xu

ECCV 2026 · Architectures and efficiency › Pruning and efficient tuning · approach: Low-rank compression and merging · 0 citations (OpenAlex)

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 Stable Rank Decomposition (adaptive residual rank) to CLIP baseline (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.832 -> 0.880Improved
add Residual Energy Constraint on top of SRD (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.880 -> 0.894Improved
add Residual Subspace Orthogonalization alone on top of SRD (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.880 -> 0.881No clear effect
add RSO on top of SRD + REC (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.894 -> 0.901Improved
replace fixed-rank Effort (best r_res=16) with adaptive stable-rank allocation (ablation)SRRA · CDF/DFDCP/DFD avgAUC: 0.897 -> 0.916Improved
replace fixed-rank LoRA (best r_res=16) with adaptive stable-rank allocation (ablation)SRRA · CDF/DFDCP/DFD avgAUC: 0.887 -> 0.916Improved
increase fixed residual rank from 16 to 64 in Effort (ablation)Effort · CDF/DFDCP/DFD avgAUC: 0.897 -> 0.891Hurt
scale backbone from CLIP ViT-B/16 to ViT-L/14 (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.849 -> 0.901Improved
use SRRA instead of Effort on ViT-B/16 (ablation)SRRA · Cross-dataset avg (ViT-B/16)AUC: 0.829 -> 0.849Improved
scale stable rank by tau=1/10 instead of using it directly (tau=1) (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.869 -> 0.901Improved
lower tau from 1/10 to 1/16 (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.901 -> 0.898Hurt
use loss weights lambda1=0.1, lambda2=0.1 instead of 0.05/0.02 (ablation)SRRA · Cross-dataset avg (train FF++ c23)AUC: 0.881 -> 0.901Improved

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