@article{dispoenhancingtrainingefficiencyandstabi,
  title = {DISPO: Enhancing Training Efficiency and Stability in Reinforcement Learning for Large Language Model Mathematical Reasoning},
  author = {Batuhan K. Karaman and Aditya Rawal and Suhaila Shakiah and Mohammad Ghavamzadeh and Mingyi Hong and Arijit Biswas and Ruida Zhou},
  year = {2026},
  eprint = {2602.00983},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.00983},
}