行业38
METR 试点评估聚焦单一 RL 训练
原标题:The pilot’s scope is restricted to the single RL run and doesn’t cover the misalignment incidents that OpenAI has disclosed so far. This ...
该试点的范围仅限于单次 RL 训练,不涵盖 OpenAI 迄今已披露的失准事件。这比我们此前提出的范围更窄,但我们希望它能提供一些关于对齐的新证据。 https://x.com/METR_Evals/status/2110519544875983177
中文译文尚未完成,先为你展示原文。
The pilot’s scope is restricted to the single RL run and doesn’t cover the misalignment incidents that OpenAI has disclosed so far. This is a narrower scope than we've previously proposed, but we hope that it will provide some new evidence about alignment.
https://x.com/METR_Evals/status/2110519544875983177
(1) Alignment The public should know if agents would attempt actions like those seen in incidents like the hack of Hugging Face. Our proposal is to analyze a developer’s internal incidents and whether training may suppress evidence of worse misalignment. https://metr.org/blog/2026-10-14-alignment-assessment-proposal/查看原帖
同一件事的其他报道1 篇
METR · 原文
