PRADA uses probability ratios of autoregressive token sequences to detect and attribute images to specific generative models.
Towards uni- versal fake image detectors that generalize across generative models
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2verdicts
UNVERDICTED 2representative citing papers
IncreFA uses hierarchical constraints with learnable orthogonal priors and a latent memory bank to enable continual adaptation for attributing images to new generative models, reporting SOTA accuracy and 98.93% unseen detection on a 28-model benchmark.
citing papers explorer
-
PRADA: Probability-Ratio-Based Attribution and Detection of Autoregressive-Generated Images
PRADA uses probability ratios of autoregressive token sequences to detect and attribute images to specific generative models.
-
IncreFA: Breaking the Static Wall of Generative Model Attribution
IncreFA uses hierarchical constraints with learnable orthogonal priors and a latent memory bank to enable continual adaptation for attributing images to new generative models, reporting SOTA accuracy and 98.93% unseen detection on a 28-model benchmark.