Vision-Encoder Behavioral Fingerprints of Image-to-Image Generative Models: A Training-Paradigm-Driven Taxonomy of Six Commercial APIs 文章

ArXiv CS.CV2026-06-16NEWSen作者: Hunter Hill
查看原文 →
关系图谱

Vision-Encoder Behavioral Fingerprints of Image-to-Image Generative Models: A Training-Paradigm-Driven Taxonomy of Six Commercial APIs · 相关技术

相关技术

DINOv2 ViT-B/14content-adaptive sub-JND adversarial perturbation pipeline