Incomplete Tasks Induce Shutdown Resistance in Some Frontier LLMs

Documenti analoghi

• Frontier LLMs Still Struggle with Simple Reasoning Tasks
  di: Malek, Alan, et al.
  Pubblicazione: (2025)
• Quantifying and Mitigating Premature Closure in Frontier LLMs
  di: Handler, Rebecca, et al.
  Pubblicazione: (2026)
• MultiChallenge: A Realistic Multi-Turn Conversation Evaluation Benchmark Challenging to Frontier LLMs
  di: Sirdeshmukh, Ved, et al.
  Pubblicazione: (2025)
• Forecasting Frontier Language Model Agent Capabilities
  di: Pimpale, Govind, et al.
  Pubblicazione: (2025)
• Scalable Qualitative Coding with LLMs: Chain-of-Thought Reasoning Matches Human Performance in Some Hermeneutic Tasks
  di: Dunivin, Zackary Okun
  Pubblicazione: (2024)