AI Paper Atlas

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

DeepFake-Adapter: Dual-Level Adapter for DeepFake Detection

Rui Shao, Tianxing Wu, Liqiang Nie, Ziwei Liu

IJCV 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Correspondence and causal representation learning · 59 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 GBA to ViT with LSA (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 73.16 -> 79.51Improved
add LSA to ViT with GBA (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 75.03 -> 79.51Improved
add LSA to ViT with GBA (ablation)DeepFake-Adapter · FF++ c40 train FS, test DFAUC: 84.90 -> 88.57Improved
add either adapter to linear-probed ViT (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 67.20 -> 73.16Improved
use parameter-efficient dual adapters instead of full-tuningDeepFake-Adapter vs full-tuning · FF++ c40 train DF, test FSAUC: 74.25 -> 76.85Improved
use dual-level adapters instead of ViT-AdapterDeepFake-Adapter vs ViT-Adapter · FF++ c40 train FS, test DFAUC: 87.65 -> 88.57Improved
supervised ImageNet-21K weights instead of MAE self-supervised weights (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 73.12 -> 79.51Improved
3 stages with 4 blocks instead of 2 stages with 6 blocks (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 75.44 -> 79.51Improved
3 stages with 4 blocks instead of 6 stages with 2 blocks (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 78.69 -> 79.51Improved
scale backbone from ViT-Base to ViT-Large (ablation)DeepFake-Adapter · FF++ c40 train DF, test FSAUC: 79.51 -> 78.60No clear effect
scale backbone from ViT-Base to ViT-Large (ablation)DeepFake-Adapter · FF++ c40 train FS, test DFAUC: 88.57 -> 85.08Hurt
use Ours* variant for cross-manipulation generalization trained with F2FDeepFake-Adapter · FF++ c40 train F2F, average over DF/FS/NTAUC: 70.29 -> 71.14Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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