ReST-RL: Achieving Accurate Code Reasoning of LLMs with Optimized Self-Training and Decoding

Documenti analoghi

• ReST-MCTS*: LLM Self-Training via Process Reward Guided Tree Search
  di: Zhang, Dan, et al.
  Pubblicazione: (2024)
• DataSciBench: An LLM Agent Benchmark for Data Science
  di: Zhang, Dan, et al.
  Pubblicazione: (2025)
• The Two-Stage Decision-Sampling Hypothesis: Understanding the Emergence of Self-Reflection in RL-Trained LLMs
  di: Zhao, Zibo, et al.
  Pubblicazione: (2026)
• SortedRL: Accelerating RL Training for LLMs through Online Length-Aware Scheduling
  di: Zhang, Yiqi, et al.
  Pubblicazione: (2026)
• Scaling Up RL: Unlocking Diverse Reasoning in LLMs via Prolonged Training
  di: Liu, Mingjie, et al.
  Pubblicazione: (2025)