LLMs are Single-threaded Reasoners: Demystifying the Working Mechanism of Soft Thinking

Documenti analoghi

• Emergent Hierarchical Reasoning in LLMs through Reinforcement Learning
  di: Wang, Haozhe, et al.
  Pubblicazione: (2025)
• Demystifying Hybrid Thinking: Can LLMs Truly Switch Between Think and No-Think?
  di: Wang, Shouren, et al.
  Pubblicazione: (2025)
• Soft Thinking: Unlocking the Reasoning Potential of LLMs in Continuous Concept Space
  di: Zhang, Zhen, et al.
  Pubblicazione: (2025)
• Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning
  di: Qian, Chen, et al.
  Pubblicazione: (2025)
• Are LLMs Capable of Data-based Statistical and Causal Reasoning? Benchmarking Advanced Quantitative Reasoning with Data
  di: Liu, Xiao, et al.
  Pubblicazione: (2024)