AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Attribution as Retrieval: Model-Agnostic AI-Generated Image Attribution

Hongsong Wang, Renxi Cheng, Chaolei Han, Jie Gui

CVPR 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
Use low-bit generative fingerprints instead of raw RGB input (ablation)ResNet-50 · GenImage 1-shot attributionmAP: +10.6% mAPImproved
Add unsupervised pretext pre-training on real-image fingerprints (ablation)LIDA · GenImage 1-shot attributionmAP: +1.5% mAPImproved
Add few-shot adaptation with center-based attribution loss (ablation)LIDA · GenImage 1-shot attributionmAP: +3.7%Improved
Add real-prototype contrastive Deepfake detection loss (ablation)LIDA · GenImage 1-shot attributionmAP: +8.2%Improved
Replace center loss LA with cross-entropy (ablation)LIDA · GenImage 1-shot attributionmAP: -1.8%Hurt
Replace detection loss LD with cross-entropy (ablation)LIDA · GenImage 1-shot attributionmAP: -0.8%Hurt
Replace both losses with cross-entropy (ablation)LIDA · GenImage 1-shot attributionmAP: -3.9%Hurt
Increase detection loss weight lambda beyond 0.9 (ablation)LIDA · GenImage attributionRank-1, mAPHurt
Increase registered shots from 1 to 10 per generatorLIDA · GenImage cross-architectureRank-1 vs mAP: Rank-1 +11.7%Mixed
Use LIDA features instead of ResNet/DIRE/ESSPLIDA · GenImage 10-shot attributionRank-1: +32.6%, +36.8%, +31.6%Improved
Use LIDA instead of FSD for 10-shot detectionLIDA · GenImage detectionaccuracy: 84.1 -> 88.3Improved
Merge diffusion generators into one paradigm classLIDA · GenImage cross-generatorRank-1, mAP: +30% Rank-1, +20% mAPImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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