UnsafeBench: Benchmarking Image Safety Classifiers on Real-World and AI-Generated Images

arXiv:2405.03486v1 Announce Type: cross
Abstract: Image safety classifiers play an important role in identifying and mitigating the spread of unsafe images online (e.g., images including violence, hateful rhetoric, etc.). At the same time, with the advent of text-to-image models and increasing concerns about the safety of AI models, developers are increasingly relying on image safety classifiers to safeguard their models. Yet, the performance of current image safety classifiers remains unknown for real-world and AI-generated images. To bridge this research gap, …

abstract ai-generated images ai models arxiv benchmarking classifiers concerns cs.cr cs.cv cs.si developers etc generated image images role safety text text-to-image type violence world