作者位次 #7/17ACL 2026Poster· Natural Language Processing
Scaling Behaviors of LLM Reinforcement Learning Post-Training: An Empirical Study in Mathematical Reasoning
Zelin Tan · Hejia Geng · Xiaohang Yu · Mulei Zhang · Guancheng Wan · Yifan Zhou · Qiang He · Xiangyuan Xue · Heng Zhou · Yutao Fan · Zhong-Zhi Li · Zaibin Zhang · Guibin Zhang · Chen Zhang · Zhenfei Yin · Philip Torr · LEI BAI
摘要
While scaling laws for large language models (LLMs) during pre-training have been extensively studied, their behavior under reinforcement learning (RL) post-training remains largely unexplored. This paper investigates the scaling behavior of Large Language Model (LLM) reinforcement learning post-...