Why Safeguarded Ships Run Aground? Aligned Large Language Models' Safety Mechanisms Tend to Be Anchored in The Template Region

Ähnliche Einträge

• No Two Devils Alike: Unveiling Distinct Mechanisms of Fine-tuning Attacks
  von: Leong, Chak Tou, et al.
  Veröffentlicht: (2024)
• Refusal Falls off a Cliff: How Safety Alignment Fails in Reasoning?
  von: Yin, Qingyu, et al.
  Veröffentlicht: (2025)
• AgentAlign: Navigating Safety Alignment in the Shift from Informative to Agentic Large Language Models
  von: Zhang, Jinchuan, et al.
  Veröffentlicht: (2025)
• Imitate Before Detect: Aligning Machine Stylistic Preference for Machine-Revised Text Detection
  von: Chen, Jiaqi, et al.
  Veröffentlicht: (2024)
• CTRAP: Embedding Collapse Trap to Safeguard Large Language Models from Harmful Fine-Tuning
  von: Yi, Biao, et al.
  Veröffentlicht: (2025)