AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Unveiling Perceptual Artifacts: A Fine-Grained Benchmark for Interpretable AI-Generated Image Detection

Yao Xiao, Weiyan Chen, Jiahao Chen, Zijie Cao, Weijian Deng, Binbin Yang, ZiYi Dong, Xiangyang Ji, Wei Ke, Pengxu Wei, Liang Lin

ICLR 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
add PAD pretraining with linear probing instead of AJ training (ablation)Swin-B · X-AIGD test AJbalanced accuracy (%): 89.3 -> 82.1Hurt
add PAD pretraining then full fine-tuning (ablation)Swin-B AJ model · X-AIGD test AJF1 (%): 90.2 -> 92.5Improved
add PAD as joint auxiliary task (multi-task) (ablation)Swin-B AJ model · X-AIGD test AJF1 (%): 90.2 -> 92.3Improved
add PAD as joint auxiliary task (multi-task) (ablation)Swin-B AJ model · X-AIGD test AJbalanced accuracy (%): 89.3 -> 89.1No clear effect
add multi-task training vs PAD-only for PAD (ablation)Swin-B · X-AIGD test category-agnostic PADIoU (%): 27.2 -> 27.3No clear effect
train on artifact annotations (PAD-only) vs existing detector heatmapsSwin-B · X-AIGD test category-agnostic PADIoU (%): 9.0 -> 27.2Improved
add attention alignment loss to DINOv2 fine-tuning (ablation)DINOv2 ViT-B · Synthbuster, CommFor, Chameleon, X-AIGDF1Improved
replace artifact masks with saliency masks in attention alignment (ablation)DINOv2 ViT-B · cross-dataset AIGI detectionF1No clear effect
raise benign region weight lambda toward 1.0 (ablation)DINOv2 ViT-B · cross-dataset AIGI detectionF1Mixed
use multi-task model for high-level and cognitive artifacts (ablation)Swin-B · X-AIGD test PAD (commonsense, physics)pixel recall (%): 2.2 -> 2.2 (commonsense); 0.3 -> 0.8 (physics)No clear effect
use general MLLM for artifact presence detectionMLLMs · X-AIGD test artifact presenceimage-level F1 (%)No clear effect

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