Curriculum Design for Trajectory-Constrained Agent: Compressing Chain-of-Thought Tokens in LLMs

Documenti analoghi

• Proximal Curriculum with Task Correlations for Deep Reinforcement Learning
  di: Tzannetos, Georgios, et al.
  Pubblicazione: (2024)
• Reward Model Learning vs. Direct Policy Optimization: A Comparative Analysis of Learning from Human Preferences
  di: Nika, Andi, et al.
  Pubblicazione: (2024)
• Informativeness of Reward Functions in Reinforcement Learning
  di: Devidze, Rati, et al.
  Pubblicazione: (2024)
• Learning Embeddings for Sequential Tasks Using Population of Agents
  di: Mahajan, Mridul, et al.
  Pubblicazione: (2023)
• Neural Task Synthesis for Visual Programming
  di: Pădurean, Victor-Alexandru, et al.
  Pubblicazione: (2023)