SDD: Self-Degraded Defense against Malicious Fine-tuning

Documenti analoghi

• CrossGuard: Safeguarding MLLMs against Joint-Modal Implicit Malicious Attacks
  di: Zhang, Xu, et al.
  Pubblicazione: (2025)
• Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning
  di: Huang, Tiansheng, et al.
  Pubblicazione: (2024)
• SOFT: Selective Data Obfuscation for Protecting LLM Fine-tuning against Membership Inference Attacks
  di: Zhang, Kaiyuan, et al.
  Pubblicazione: (2025)
• Real-time ML-based Defense Against Malicious Payload in Reconfigurable Embedded Systems
  di: Stahle-Smith, Rye, et al.
  Pubblicazione: (2025)
• Harmful Fine-tuning Attacks and Defenses for Large Language Models: A Survey
  di: Huang, Tiansheng, et al.
  Pubblicazione: (2024)