@article{sgrpoearlyexitviareinforcementlearning,
  title = {S-GRPO: Early Exit via Reinforcement Learning in Reasoning Models},
  author = {Muzhi Dai and Chenxu Yang and Qingyi Si},
  year = {2025},
  eprint = {2505.07686},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2505.07686v1},
}