The Dual Power of Interpretable Token Embeddings: Jailbreaking Attacks and Defenses for Diffusion Model Unlearning

Documenti analoghi

• Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models
  di: Zhang, Yimeng, et al.
  Pubblicazione: (2024)
• To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images ... For Now
  di: Zhang, Yimeng, et al.
  Pubblicazione: (2023)
• UnlearnCanvas: Stylized Image Dataset for Enhanced Machine Unlearning Evaluation in Diffusion Models
  di: Zhang, Yihua, et al.
  Pubblicazione: (2024)
• The VLLM Safety Paradox: Dual Ease in Jailbreak Attack and Defense
  di: Guo, Yangyang, et al.
  Pubblicazione: (2024)
• Jailbreak Attacks and Defenses against Multimodal Generative Models: A Survey
  di: Liu, Xuannan, et al.
  Pubblicazione: (2024)