New paper on robustness evalutation of unlearned image generation models

[ryliu68]

Hi, and thanks for maintaining this project.

I’d like to share a new paper that is relevant to robustness evaluation of unlearned image generation models:

• Title: "Image Can Bring Your Memory Back: A Novel Multi-Modal Guided Attack against Image Generation Model Unlearning" (arXiv:2507.07139, 2025)
• Authors: Renyang Liu, Guanlin Li, Tianwei Zhang, See-Kiong Ng

Why it may be relevant: The paper proposes a multi-modal guided attack that uses an image to recover “forgotten” concepts from unlearned IGMs, which can stress-test unlearning robustness.

If helpful, please consider adding it to the list. Thank you very much!