1
Voting with the Graph: Stable RLAIF via Topological Consistency Maximization
用图投票和拓扑一致性最大化让RLAIF训练更稳定，清华等团队提出新对齐方法。
arXiv:2510.15514v3 Announce Type: replace Abstract: Reinforcement Learning from AI Feedback (RLAIF) relies on LLM judges as preference measurement ins…