@article{sbtrpotowardssafereinforcementlearningwi,
  title = {SB-TRPO: Towards Safe Reinforcement Learning with Hard Constraints},
  author = {Dominik Wagner and Ankit Kanwar and Luke Ong},
  year = {2025},
  eprint = {2512.23770},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2512.23770},
}