AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Leveraging Failed Samples: A Few-Shot and Training-Free Framework for Generalized Deepfake Detection

Shibo Yao, Renshuai Tao, Xiaolong Zheng, Chao Liang, Chunjie Zhang

AAAI 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot · 1 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
fine-tune the cache keys (FTNet-T) instead of training-free FTNet (ablation)FTNet · UniversalFakeDetectmAcc: 94.03 -> 97.14Improved
fine-tune the cache keys (FTNet-T) instead of training-free FTNet (ablation)FTNet · GenImage cross-domainmAcc: 90.7 -> 94.2Improved
fine-tune the cache keys (FTNet-T) instead of training-free FTNet (ablation)FTNet · OpenSDImAcc: 79.94 -> 83.16Improved
use 8-shot fine-tuned baselines UniFD/NPR vs FTNet-T with only 4 shotsFTNet-T · GenImage, UniFD, OpenSDI averagemAcc: 69.3 -> 90.3Improved
FTNet-T trained on 500 images vs baselines trained on 320k SDv1.4 imagesFTNet-T · GenImage zero-shot generalizationmAcc: 84.6 -> 86.0Improved
training-free FTNet with 500 images vs UniFD trained on 320k imagesFTNet · GenImage zero-shot generalizationmAcc: 84.6 -> 80.8Hurt
use more shots in the cache (ablation)FTNet · GenImage, UniFD, OpenSDIaccuracyMixed
use CLIP middle layer L12 instead of L6, L18 or L24 features (ablation)FTNet · GenImage 4-shotaverage accuracyImproved
use ViT-L/14 backbone instead of ViT-B/16 (ablation)FTNet · GenImageaccuracy: 60.4 -> 90.7Improved
use ViT-L/14 backbone instead of ViT-B/32 (ablation)FTNet · GenImageaccuracy: 61.2 -> 90.7Improved
use ViT-L/14 backbone instead of ViT-B/16 (ablation)FTNet · Chameleonaccuracy: 77.4 -> 80.7Improved
apply JPEG compression and Gaussian blur to test imagesFTNet · ChameleonaccuracyHurt

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