Intrinsic Self-correction for Enhanced Morality: An Analysis of Internal Mechanisms and the Superficial Hypothesis

Documenti analoghi

• On the Intrinsic Self-Correction Capability of LLMs: Uncertainty and Latent Concept
  di: Liu, Guangliang, et al.
  Pubblicazione: (2024)
• A Survey to Recent Progress Towards Understanding In-Context Learning
  di: Mao, Haitao, et al.
  Pubblicazione: (2024)
• On the Convergence of Moral Self-Correction in Large Language Models
  di: Liu, Guangliang, et al.
  Pubblicazione: (2025)
• Discourse Heuristics For Paradoxically Moral Self-Correction
  di: Liu, Guangliang, et al.
  Pubblicazione: (2025)
• Self-correction is Not An Innate Capability in Language Models
  di: Liu, Guangliang, et al.
  Pubblicazione: (2024)