RL Fine-Tuning Heals OOD Forgetting in SFT

@misc{jinSat Nov 01 2025 13:50:29 GMT+0000 (Coordinated Universal Time)rlfinetuningheals,
      title={RL Fine-Tuning Heals OOD Forgetting in SFT},
      author={Hangzhan Jin and Sitao Luan and Sicheng Lyu and Guillaume Rabusseau and Reihaneh Rabbany and Doina Precup and Mohammad Hamdaqa},
      year={Sat Nov 01 2025 13:50:29 GMT+0000 (Coordinated Universal Time)},
      eprint={2509.12235},
      archivePrefix={arXiv},
      primaryClass={cs.AI},
      url={https://arxiv.org/abs/2509.12235},
}