On the Overscaling Curse of Parallel Thinking: System Efficacy Contradicts Sample Efficiency

Documenti analoghi

• Embedding Trajectory for Out-of-Distribution Detection in Mathematical Reasoning
  di: Wang, Yiming, et al.
  Pubblicazione: (2024)
• An Analysis and Mitigation of the Reversal Curse
  di: Lv, Ang, et al.
  Pubblicazione: (2023)
• Mitigating Reversal Curse in Large Language Models via Semantic-aware Permutation Training
  di: Guo, Qingyan, et al.
  Pubblicazione: (2024)
• The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse and More
  di: Kitouni, Ouail, et al.
  Pubblicazione: (2024)
• Demystifying Hybrid Thinking: Can LLMs Truly Switch Between Think and No-Think?
  di: Wang, Shouren, et al.
  Pubblicazione: (2025)