@article{noisecorrectedgrpofromnoisyrewardstounbi, title = {Noise-corrected GRPO: From Noisy Rewards to Unbiased Gradients}, author = {Omar El Mansouri and Fathinah Asma Izzati and Mohamed El Amine Seddik and Salem Lahlou}, year = {2025}, eprint = {2510.18924}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2510.18924}, }