UPAM: Unified Prompt Attack in Text-to-Image Generation Models Against Both Textual Filters and Visual Checkers.

Duo Peng, Qiuhong Ke, Jun Liu

Browse the full ICML paper archive.