@article{mitigatingrewardoveroptimizationinrlhf,
  title = {Mitigating Reward Over-Optimization in RLHF via Behavior-Supported Regularization},
  author = {Juntao Dai and Taiye Chen and Yaodong Yang and Qian Zheng and Gang Pan},
  year = {2025},
  eprint = {2503.18130},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2503.18130v1},
}