@article{vrporethinkingvaluemodelingforrobustrlun, title = {VRPO: Rethinking Value Modeling for Robust RL under Noisy Supervision in LLM Post-Training}, author = {Dingwei Zhu and Shihan Dou and Zhiheng Xi and Senjie Jin and Guoqiang Zhang and Jiazheng Zhang and Junjie Ye and Mingxu Chai and Enyu Zhou and Ming Zhang and Yuhui Wang and Caishuang Huang and Chenhao Huang and Yunke Zhang and Yuran Wang and Tao Gui and Qi Zhang and Xipeng Qiu and Xuanjing Huang}, year = {2025}, eprint = {2508.03058}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2508.03058}, }