CoDA is a lightweight detector using a Noise-Quantization Probe on color non-uniformity that reports strong cross-domain results on the new FakeForm benchmark and competitive cross-model performance on standard tests.
arXiv preprint arXiv:2603.08064 (2026)
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
Automatic metrics such as FID are misaligned with human perception and downstream segmentation performance for Earth observation datasets and synthetic counterparts.
citing papers explorer
-
CoDA: Color Distribution Probing for Efficient and Generalizable AI-Generated Image Detection
CoDA is a lightweight detector using a Noise-Quantization Probe on color non-uniformity that reports strong cross-domain results on the new FakeForm benchmark and competitive cross-model performance on standard tests.
-
Benchmarking the Alignment of Data-Quality Metrics, Human Judgment and Land-Cover Segmentation Performance for Earth Observation
Automatic metrics such as FID are misaligned with human perception and downstream segmentation performance for Earth observation datasets and synthetic counterparts.