AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Prior-Conditioned Gaussian Discriminants for Generalizable AI-generated Image Detection

Shashank Kotyan, Makoto Shing, Yuki Imajuku, Rujikorn Charakorn, Tarin Clanuwat

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 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
replace released head with Mah-NCM on same featuresUnivFD · Percept-Lens, CNNSpot priormean class accuracy (%): 54.63 -> 57.62Improved
replace released head with Mah-NCM on same featuresAIDE (Full GenImage) · Percept-Lens, GenImage priormean class accuracy (%): 54.98 -> 64.64Improved
replace released head with Mah-NCM on same featuresDRCT-UnivFD (Full GenImage) · Percept-Lens, GenImage priormean class accuracy (%): 65.77 -> 73.01Improved
replace released head with QDA on same featuresDRCT-UnivFD (DRCT-SDv2) · Percept-Lens, DRCT priormean class accuracy (%): 62.37 -> 70.31Improved
replace released head with Mah-NCM on same featuresCF-384 · Percept-Lens, CF priormean class accuracy (%): 87.54 -> 84.55Hurt
use shared full covariance (Mah-NCM) instead of isotropic Euc-NCM (ablation)PE-Core-bigG-14-448 head · Percept-Lens, CF priormean class accuracy (%): 86.38 -> 94.46Improved
use shared covariance (Mah-NCM) instead of diagonal GNB (ablation)DINOv3-ViT-7b head · Percept-Lens, CF priormean class accuracy (%): 83.18 -> 88.91Improved
use class-specific covariance (QDA) instead of Mah-NCM (ablation)PE-Core-bigG-14-448 head · Percept-Lens, CF priormean class accuracy (%): 94.46 -> 86.15Hurt
use class-specific covariance (QDA) instead of Mah-NCM (ablation)PE-Core-bigG-14-448 head · Percept-Lens, DRCT priormean class accuracy (%): 83.25 -> 91.09Improved
change training prior from CNNSpot to CF (ablation)PE-Core-bigG-14-448 Gaussian head · Percept-Lensmean class accuracy (%): 77.92 -> 94.46Improved
scale encoder from CLIP ResNet-50 to ViT-L/14 (ablation)Mah-NCM head · Percept-Lens, CF priormean class accuracy (%): 72.00 -> 76.96Improved
change pretraining objective from MAE-Huge to DINOv3-ViT-7b (ablation)Mah-NCM head · Percept-Lens, CF priormean class accuracy (%): 69.37 -> 88.91Improved

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