Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection

Documenti analoghi

• Refusal Before Decoding: Detecting and Exploiting Refusal Signals in Intermediate LLM Activations
  di: Collu, Matteo Gioele, et al.
  Pubblicazione: (2026)
• Gradient Cuff: Detecting Jailbreak Attacks on Large Language Models by Exploring Refusal Loss Landscapes
  di: Hu, Xiaomeng, et al.
  Pubblicazione: (2024)
• From Threat to Tool: Leveraging Refusal-Aware Injection Attacks for Safety Alignment
  di: Chae, Kyubyung, et al.
  Pubblicazione: (2025)
• A Content-Based Framework for Cybersecurity Refusal Decisions in Large Language Models
  di: Linder, Noa, et al.
  Pubblicazione: (2026)
• Why Do Aligned LLMs Remain Jailbreakable: Refusal-Escape Directions, Operator-Level Sources, and Safety-Utility Trade-off
  di: Chen, Yu, et al.
  Pubblicazione: (2026)