Noise Injection Systemically Degrades Large Language Model Safety Guardrails

Documenti analoghi

• Where Norms and References Collide: Evaluating LLMs on Normative Reasoning
  di: Abrams, Mitchell, et al.
  Pubblicazione: (2026)
• Why Do Safety Guardrails Degrade Across Languages?
  di: Zhang, Max, et al.
  Pubblicazione: (2026)
• UniGuard: Towards Universal Safety Guardrails for Jailbreak Attacks on Multimodal Large Language Models
  di: Oh, Sejoon, et al.
  Pubblicazione: (2024)
• Identifying Features Associated with Bias Against 93 Stigmatized Groups in Language Models and Guardrail Model Safety Mitigation
  di: Gueorguieva, Anna-Maria, et al.
  Pubblicazione: (2025)
• Probing a Vision-Language-Action Model for Symbolic States and Integration into a Cognitive Architecture
  di: Lu, Hong, et al.
  Pubblicazione: (2025)