1
Bayesian Inference and Decision Audits for Public Archives of Frontier AI Evaluations
用贝叶斯推断重构前沿AI评估档案，打破“排行榜”式解读，推动更科学、透明的决策审计。
arXiv:2606.17005v1 Announce Type: new Abstract: Public AI evaluations are often read as terminal leaderboards, yet the underlying evidence is a select…