1
Label-Free Reinforcement Learning via Cross-Model Entropy
提出Cross-Model Entropy方法，免去人工标注，以模型间交叉熵驱动强化学习后训练，突破奖励信号瓶颈。
arXiv:2605.29009v1 Announce Type: cross Abstract: Post-training large language models with reinforcement learning is bottlenecked by the reward signal…