@inproceedings{Zhou2026Beyond,
  title={Beyond the Stability-Exploration Dilemma: Environmental Regularization for LLM Policy Optimization},
  author={Zhou, Xianlei and Meng, Xiangdi and He, Yu and Qi, Tianyu and Guan, Shuyan and Zhang, Xianli and Zhang, Jian and Li, Xin and Lin, Qika and Liu, Jun},
  booktitle={Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year={2026},
  note={Accepted}
}