📖 Anti-Dreambooth
Cite as: ICCV 2023 Submitted on 2023/10 Algorithm
Jolie's Website
📖 Advdm vs Anti-DB
目標差異 防禦重點不同： GLAZE 和 AdvDM：主要專注於防止藝術風格模仿（artistic mimicry），目的是保護藝術家的創作風格不被 AI 模型複製。 Anti-DreamBooth：專注於防止生成假冒或有害的個人化圖像，更關注保護個人隱私和防止身份濫用的負面影響。 技術方法差異 實現機制不同： GLAZE：...
📖 DisDiff (Disrupting Diffusion)+Advdm
Liu et al. 2024. ACM International Conference on Multimedia (MM ’24) REF: Disrupting Diffusion: Token-Level Attention Erasure Attack against Diffusion-based Customization Review about protect me...
📖 Progress+gpt4o-img-gen
Survey Safety at Scale: A Comprehensive Survey of Large Model Safety Xingjun Ma, Yifeng Gao, Yixu Wang, Ruofan Wang, Xin Wang, Ye Sun, … ArXiv:2502.05206 Submitted on 2025/03 Intellectual Prop...
📖 MetaCloak_ Preventing Unauthorized Subject-driven Text-to-image Diffusion-based Synthesis via Meta-learning
Cite as: CVPR Submitted on 2024 Yixin Liu, Chenrui Fan, Yutong Dai, Xun Chen, Pan Zhou, Lichao Sun Abstract Problem: Text-to-image models can create harmful content, risking individual safety. ...
📖 Glaze_ Protecting Artists from Style Mimicry by Text-to-Image Models
Shawn Shan, Jenna Cryan, Emily Wenger, Haitao Zheng, Rana Hanocka, Ben Y. Zhao Cite as: USENIX Security 2023 Submitted on 2023/08 Abstract Models can learn to mimic the artistic style of specif...
📖 Towards Prompt-robust Face Privacy Protection via Adversarial Decoupling Augmentation Framework
Cite as: arXiv:2305.03980 [cs.CV] Submitted on 2023/05 Ruijia Wu, Yuhang Wang, Huafeng Shi, Zhipeng Yu, Yichao Wu, Ding Liang Abstract The image-text fusion module is a vital component of text-...
📖 MalImgDA_ Diffusion-based Data Augmentation for Long-tailed Malware Family Classification
Gang Yang; Jun He; Bo Wu; Tao Xia; Linna Fan; Lin Ni Cite as: IEEE ICASSP 2025 Submitted on 2025/03 Abstract Novel Framework: The paper proposes a new data augmentation framework using a diffus...
📖 Training Diffusion Models with Federated Learning
Matthijs de Goede, Bart Cox, Jérémie Decouchant Cite as: arXiv:2406.12575 [cs.LG] Submitted on 18 Jun 2024 Abstract Raising concerns about privacy, copyright, and data authority due to their la...
📖 Weekly Progress - Paper Reading & Survey
Paper #1: MalImgDA MalImgDA: Diffusion-based Data Augmentation for Long-tailed Malware Family Classification Gang Yang; Jun He; Bo Wu; Tao Xia; Linna Fan; Lin Ni 2025 IEEE International Confer...