@article{xiong2026selfsteprejection,
  title={Can Language Models Learn to Reject Their Own Bad Reasoning Steps?},
  author={Xiong, Siheng and Liu, Xiaoze and Jin, Yiqiao and Wang, Xiaoqian and Gao, Jing},
  journal={arXiv preprint arXiv:2610.05976},
  year={2026},
  url={https://arxiv.org/abs/2610.05976}
}