AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

Zero-shot Detection of AI-Generated Image via RAW-RGB Alignment

Haiwei Wu, Fengpeng Li, Zhilin Tu, Yuanman Li, Xiong Li, Jiantao Zhou

CVPR 2026 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Transductive and semantic-dictionary zero-shot

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
align RGB features with RAW-derived pipeline graphs and RAW latents, trained only on real RAW-RGB pairszero-shot AIGC detector · Zero-shot clustering avg over GANs and DMsNMI: .742 -> .964 vs MIBImproved
use RAW-RGB alignment trace instead of baseline features for pair similarityzero-shot AIGC detector · Feature similarity avgAUC: .734 -> .925 vs MIBImproved
fine-tune the alignment feature space on ProGAN onlyzero-shot AIGC detector · ForenSynths cross-domainAP: .953 -> .987 vs EffortImproved
freeze the pre-trained ViT instead of training it (ablation)RGB-Vision branch · Ablation, detectionAP: .909 -> .963Improved
replace standard cross-attention with Stratified Entropy Attention (ablation)RGB-Vision branch · Ablation, detectionAP: .969 -> .987Improved
use ViT backbone instead of ConvNeXt, EfficientNet or SegFormer (ablation)RGB-Vision branch · Ablation, detectionAP: ConvNeXt .882, SegFormer .875 -> ViT (frozen) .963Improved
use all ISP plus post-processing pipelines instead of ISP-only or post-only (ablation)RAW-Graph branch · Ablation, detectionAP: .843 / .858 -> .885Improved
encode pipelines as directed weighted graph instead of text or undirected graph (ablation)RAW-Graph branch · Ablation, detectionAP: .885 text -> .917 undirected -> .958 directed -> .987 weightedImproved
use InfoNCE instead of Triplet or Circle loss (ablation)RAW-Graph branch · Ablation, detectionAP: .924 / .965 -> .987Improved
add RAW-Vision branch (ablation)Alignment trace model · Ablation, detectionAP: .896 -> .987Improved
use VAE with KL constraint instead of UNet with L2 (ablation)RAW-Vision branch · Ablation, detectionAP: .908 -> .975Improved
add cross-attention conditioning on trace feature in the RAW-Vision branch (ablation)RAW-Vision branch · Ablation, detectionAP: .975 -> .987Improved

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