Z. An and W. Du, “MoralReason: Generalizable Moral Decision Alignment for LLM Agents Using Reasoning-Level Reinforcement Learning”, AAAI, vol. 40, no. 44, pp. 37232–37239, Mar. 2026.