@article{
yu2026reassessing,
title={Reassessing the Role of Supervised Fine-Tuning and Reinforcement Learning: An Empirical Study in {VLM} Reasoning},
author={Yongcan Yu and Lingxiao He and Shuo Lu and Lijun Sheng and Yinuo Xu and Yanbo Wang and Kuangpu Guo and Jianjie Cheng and Meng Wang and Qianlong Xie and Xingxing Wang and Dapeng Hu and Jian Liang},
journal={Transactions on Machine Learning Research},
issn={2835-8856},
year={2026},
url={https://openreview.net/forum?id=EWk0VxDpax},
note={}
}