Lu, S.; Wang, H.; Chen, Z.; Tang, Y. URPO: A Unified Reward & Policy Optimization Framework for Large Language Models. AAAI 2026, 40, 32329-32337.