IStego100K++: A Large-Scale GAN-Based Steganalysis Dataset Using Adversarially Learned Embedding Costs
Steganalysis benchmarks built with fixed distortion functions — such as nsF5, J-UNIWARD, and UERD — are increasingly inadequate against GAN-based steganography, where adversarially learned embedding costs bypass traditional rich-model detectors. To address this gap, we introduce IStego100K++, a large-scale GAN-generate...