When Refusals Fail: Unstable Safety Mechanisms in Long-Context LLM Agents

Documenti analoghi

• Evaluation of Few-Shot Learning for Classification Tasks in the Polish Language
  di: Hadeliya, Tsimur, et al.
  Pubblicazione: (2024)
• Detection Is Cheap, Routing Is Learned: Why Refusal-Based Alignment Evaluation Fails
  di: Frank, Gregory N.
  Pubblicazione: (2026)
• LongSafety: Enhance Safety for Long-Context LLMs
  di: Huang, Mianqiu, et al.
  Pubblicazione: (2024)
• PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents
  di: Gu, Zhuohan, et al.
  Pubblicazione: (2026)
• When LLM Judge Scores Look Good but Best-of-N Decisions Fail
  di: Landesberg, Eddie
  Pubblicazione: (2026)