1
MemReward: Graph-Based Experience Memory for LLM Reward Prediction with Limited Labels
用图结构记忆经验，破解LLM奖励预测中标签不足的难题，创新方法值得关注。
arXiv:2603.19310v3 Announce Type: replace-cross Abstract: Reinforcement learning has emerged as a powerful paradigm for improving large language model…