@article{stabilizingpolicygradientmethodsviarewar, title = {Stabilizing Policy Gradient Methods via Reward Profiling}, author = {Shihab Ahmed and El Houcine Bergou and Aritra Dutta and Yue Wang}, year = {2025}, eprint = {2511.16629}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2511.16629}, }