An, Z.; Du, W. MoralReason: Generalizable Moral Decision Alignment for LLM Agents Using Reasoning-Level Reinforcement Learning. AAAI 2026, 40, 37232-37239.