@article{pureexplorationforagoodpolicyinreinforce,
  title = {Pure Exploration for a Good Policy in Reinforcement Learning with Bandit Feedback},
  author = {Zitian Li and Wang Chi Cheung},
  year = {2026},
  eprint = {2605.23182},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.23182},
}