Eliciting Chain-of-Thought in Base LLMs via Gradient-Based Representation Optimization

Documenti analoghi

• ThoughtProbe: Classifier-Guided LLM Thought Space Exploration via Probing Representations
  di: Wang, Zijian, et al.
  Pubblicazione: (2025)
• ThoughtProbe: Classifier-Guided Thought Space Exploration Leveraging LLM Intrinsic Reasoning
  di: Wang, Zijian, et al.
  Pubblicazione: (2025)
• Strategic Chain-of-Thought: Guiding Accurate Reasoning in LLMs through Strategy Elicitation
  di: Wang, Yu, et al.
  Pubblicazione: (2024)
• Chain-of-Conceptual-Thought Elicits Daily Conversation in Large Language Models
  di: Gu, Qingqing, et al.
  Pubblicazione: (2025)
• Graph-Based Chain-of-Thought Pruning for Reducing Redundant Reflections in Reasoning LLMs
  di: Yuan, Hongyuan, et al.
  Pubblicazione: (2026)