Is It Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort

Ähnliche Einträge

• Think Before Refusal : Triggering Safety Reflection in LLMs to Mitigate False Refusal Behavior
  von: Si, Shengyun, et al.
  Veröffentlicht: (2025)
• Spontaneous Reward Hacking in Iterative Self-Refinement
  von: Pan, Jane, et al.
  Veröffentlicht: (2024)
• Look at the Text: Instruction-Tuned Language Models are More Robust Multiple Choice Selectors than You Think
  von: Wang, Xinpeng, et al.
  Veröffentlicht: (2024)
• LLMs Are Prone to Fallacies in Causal Inference
  von: Joshi, Nitish, et al.
  Veröffentlicht: (2024)
• Teaching Models to Verbalize Reward Hacking in Chain-of-Thought Reasoning
  von: Turpin, Miles, et al.
  Veröffentlicht: (2025)