AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

APT: Anchor-aligned Perturbations for Tamper Localization in Fully Regenerated Images

Suhyeon Ha, Woo Jae Kim, Joonsung Jeon, Sooel Son, Sung-eui Yoon

ECCV 2026 · Image generation › Autoregressive and latent generation · approach: Generalizable fake image detection · 0 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 noisy perturbation branch loss (ablation)APT · COCO FRIoU: +0.24Improved
add noisy perturbation branch loss (ablation)APT · COCO SPIoU: +0.23Improved
add hard negative mining loss (ablation)APT · COCO FRIoU: +0.26Improved
add hard negative mining loss (ablation)APT · COCO SPIoU: +0.29Improved
add shallow mask decoder to training-free predictionAPT · COCO 500 images FRIoU: 0.89 -> 0.91Improved
add shallow mask decoder under corruptionsAPT · Corruption average FRIoU: 0.80 -> 0.90Improved
use encoder layer 1 instead of layers 0, 2, 3 (ablation)APT · FR center-mask SD-Paintercosine similarity foreground/background separation: 0.124 vs 0.265Improved
reduce optimization steps from 150 to 50 (ablation)APT* · COCO FR and SPIoU: 42.52 s -> 14.34 s per imageNo clear effect
tighten perturbation budget epsilon (ablation)APT* · COCOIoU: modest localization cost for better qualityMixed
embed latent perturbation (APT) instead of pixel/frequency embeddingAPT vs StableGuard/OmniGuard · COCO FR SD-PainterAUC: 0.5 -> 0.97Improved
apply APT* to non-identical VAE and flow-matching inpaintersAPT* · FR transferIoU: IoU >= 0.91No clear effect

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 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