AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Enabling Supervised Learning of Generative Signatures for Generalized AI-Generated Images Detection

Jianwei Fei, Yunshu Dai, Xiaoyu Zhou, Zhihua Xia, Alessandro Piva

CVPR 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection

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 self-augmentation with reconstructed images (p_aug 20%) (ablation)GenSign detector · mean mAP across 3 benchmarksmAP: 88.80 -> 95.79Improved
increase reconstructor downsampling levels from 1 to 3 (ablation)DAIR · mean APAP: 76.6 -> 95.8Improved
feed GenSign instead of RGB to CNN from scratch (ablation)ResNet50 · mean mAPmAP: 68.93 -> 78.86Improved
feed GenSign instead of RGB to CNN from scratch (ablation)EfficientNet-B0 · mean mAPmAP: 80.88 -> 83.59Improved
replace RGB with GenSign input (ablation)CLIP ViT-L/14 · mean mAPmAP: 88.39 -> 87.93Hurt
add GenSign stream to RGB stream (dual-stream) (ablation)CLIP detector · mean mAPmAP: 88.39 -> 95.79Improved
use GenSign extractor instead of SRM Conv (ablation)dual-stream detector · mean mAPmAP: 95.27 -> 95.79Improved
use GenSign extractor instead of Noiseprint++ (ablation)dual-stream detector · mean mAPmAP: 86.18 -> 95.79Improved
use GenSign detector instead of FatFormerFatFormer · UniversalFakeDetectmean AP: 98.16 -> 99.37Improved
use GenSign detector instead of AIDEAIDE · AIGIBenchmarkmean AP: 75.36 -> 89.96Improved
use GenSign detector instead of EffortEffort · GenImagemean accuracy: 91.1 -> 96.6Improved
use GenSign detector instead of UniFDUniFD · AIGCDetectBenchmark: MidjourneyAP: 74.00 -> 81.17Improved

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