Wisdom is Knowing What not to Say: Hallucination-Free LLMs Unlearning via Attention Shifting

Documenti analoghi

• Less is More: Geometric Unlearning for LLMs with Minimal Data Disclosure
  di: Tan, Chenchen, et al.
  Pubblicazione: (2026)
• When Harmless Words Harm: A New Threat to LLM Safety via Conceptual Triggers
  di: Zhang, Zhaoxin, et al.
  Pubblicazione: (2025)
• Large Language Models Know What To Say But Not When To Speak
  di: Umair, Muhammad, et al.
  Pubblicazione: (2024)
• You Know What I'm Saying: Jailbreak Attack via Implicit Reference
  di: Wu, Tianyu, et al.
  Pubblicazione: (2024)
• LLMs Know More About Numbers than They Can Say
  di: Yuchi, Fengting, et al.
  Pubblicazione: (2026)