AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

FiSeR: Fine-Grained Source Representations for Cross-Domain AI Image Detection

Shan Zhang, Yongxin He, Mingming Zhang, Huiwen Tian, Lei Ma

ICML 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Cross-domain few-shot with transformers · 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
train with hierarchical (coarse + fine) contrastive learning on DINOv3 instead of DIREFiSeR vs DIRE · Average over WildFake/Community/AIGIBench/Chameleon/GenImage (trained on WildFake)AUROC: 88.01 -> 98.23Improved
add fine-grained generator-identity contrastive loss (ablation)FiSeR · Average over WildFake/Community/AIGIBench/Chameleon/GenImage (trained on WildFake)AUROC: 94.75 -> 98.23Improved
add coarse-grained natural-vs-synthetic contrastive loss (ablation)FiSeR · Average over WildFake/Community/AIGIBench/Chameleon/GenImage (trained on WildFake)AUROC: 93.64 -> 98.23Improved
add fine-grained loss (ablation)FiSeR · AIGIBench (trained on WildFake)AUROC: 88.83 -> 98.00Improved
add coarse-grained loss (ablation)FiSeR · AIGIBench (trained on WildFake)AUROC: 93.15 -> 98.00Improved
fine-tune DINOv3 ViT-L/16 on WildFake with contrastive loss instead of pretrained (ablation)FiSeR · AIGIBench (trained on WildFake)AUROC: 89.87 -> 98.00Improved
use larger DINOv3 ViT-H/16+ instead of ViT-L/16 after fine-tuning (ablation)FiSeR · AIGIBench (trained on WildFake)AUROC: 98.00 -> 97.24Hurt
use DINOv3 backbone instead of CLIP ViT-L/14 (fine-tuned) (ablation)FiSeR · AIGIBench (trained on WildFake)AUROC: 93.78 -> 98.00Improved
refit an SVM head with 10 target samples per class on frozen featuresResNet50-Detection · ChameleonAUROC: 63.65 -> 92.38Improved
refit an SVM head with 10 target samples per class on frozen featuresCLIP-Detection · ChameleonAUROC: 60.81 -> 85.39Improved
use SVM instead of k-NN head for few-shot adaptation (ablation)all methods average · AIGIBench 10-shotAUROC: 84.12 -> 87.82Improved
train on Community instead of WildFakeFiSeR · Average over 5 datasetsAUROC: 98.23 -> 91.12Hurt

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