AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

DySy-Det: A Synergistic Framework with Dynamic Reconstruction-Path Consistency for AI-Generated Image Detection

Fanli Jin, Feng Lin, Gaojian Wang, Tong Wu, Zhisheng Yan

AAAI 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 2 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 guided reconstruction error (ablation)DySy-Det · GenImage (train SD V1.4)mAcc: 91.92 -> 93.02Improved
add RPC feature (ablation)DySy-Det · GenImage (train SD V1.4)mAcc: 90.45 -> 93.02Improved
add attention mask guidance (ablation)DySy-Det · GenImage (train SD V1.4)mAcc: 91.52 -> 93.02Improved
use full three-cue DySy-Det instead of CLIP only (ablation)DySy-Det · GenImage (train SD V1.4)mAcc: 84.85 -> 93.02Improved
DySy-Det vs prior best AIDEDySy-Det · GenImage (train SD V1.4)mAcc: 86.88 -> 93.02Improved
DySy-Det vs prior best FatFormerDySy-Det · UniversalFakeDetectmAcc: 89.99 -> 91.56Improved
apply JPEG compression at quality 50DySy-Det · JPEG robustnessmAccMixed

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