1
Beyond Scalar Rewards: Dense Feedback for LLM Policy Synthesis in Sequential Social Dilemmas
提出密集反馈机制替代标量奖励，为LLM在序列社会困境中的策略合成开辟新路径。
arXiv:2603.19453v2 Announce Type: replace Abstract: We study LLM policy synthesis: using a language model to iteratively generate programmatic agent p…