@article{stairaddressingstagemisalignmentthrought,
  title = {STAIR: Addressing Stage Misalignment through Temporal-Aligned Preference Reinforcement Learning},
  author = {Yao Luan and Ni Mu and Yiqin Yang and Bo Xu and Qing-Shan Jia},
  year = {2025},
  eprint = {2509.23802},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2509.23802},
}