AI-generated image detection on Stable Diffusion generator v3.0 (test)
83.58AccuracySAFE
Evaluation results
| Method | Links | ||
|---|---|---|---|
| SAFETraining Dataset=ProGAN2026.01 | 83.58 | 94.08 | |
| NPRTraining Dataset=ProGAN2026.01 | 82.95 | 94.02 | |
| FatFormerTraining Dataset=ProGAN2026.01 | 74.49 | 89.52 | |
| Ours (ViT-L/14)Training Dataset=ProGAN, Backbone=ViT-L/142026.01 | 70.24 | 81.63 | |
| LGradTraining Dataset=ProGAN2026.01 | 70.13 | 77.89 | |
| Ours (ViT-B/32)Training Dataset=ProGAN, Backbone=ViT-B/322026.01 | 68.6 | 76.46 | |
| GramNetTraining Dataset=ProGAN2026.01 | 59.15 | 58.15 | |
| LNPTraining Dataset=ProGAN2026.01 | 57.14 | 62.15 | |
| UnivFDTraining Dataset=ProGAN2026.01 | 50.12 | 59.05 | |
| CNNSpotTraining Dataset=ProGAN2026.01 | 49.91 | 50.94 | |
| FreDectTraining Dataset=ProGAN2026.01 | 40.98 | 42.15 |