@article{learningfromfailurescorrectionorientedpo,
  title = {Learning from Failures: Correction-Oriented Policy Optimization with Verifiable Rewards},
  author = {Mengjie Ren and Jie Lou and Boxi Cao and Xueru Wen and Hongyu Lin and Xianpei Han and Le Sun and Xing Yu and Yaojie Lu},
  year = {2026},
  eprint = {2605.14539},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2605.14539},
}