arXiv cs.LG
· Papers
All-Quadrant Bounded Clipping GRPO: Closing the Unbounded Blind Spot for Stable and Generalizable Training
arXiv:2601.03895v2 Announce Type: replace Abstract: Group Relative Policy Optimization (GRPO) has emerged as a popular algorithm for reinforcement learning with large language models (LLMs). However, GRPO inherits PPO's token-level clipping while replacing token-level advantages with a single sequence-level advantage.