TY  - RPRT
TI  - REFORGE: Multi-modal Attacks Reveal Vulnerable Concept Unlearning in Image Generation Models
AU  - Yong Zou
AU  - Haoran Li
AU  - Fanxiao Li
AU  - Shenyang Wei
AU  - Yunyun Dong
AU  - Li Tang
AU  - Wei Zhou
AU  - Renyang Liu
PY  - 2026
UR  - https://arxiv.org/abs/2603.16576
ID  - 2603.16576
ER  -