@article{processrewardswithlearnedreliability, title = {Process Rewards with Learned Reliability}, author = {Jinyuan Li and Langlin Huang and Chengsong Huang and Shaoyang Xu and Donghong Cai and Yuyi Yang and Wenxuan Zhang and Jiaxin Huang}, year = {2026}, eprint = {2605.15529}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2605.15529}, }