# scalable-oversight
標記為「scalable-oversight」的 2 篇文章
Scalable Oversight Challenges
How oversight breaks down as AI systems become more capable: the scalable oversight problem, recursive reward modeling, debate, market-making, and implications for red teaming increasingly capable models.
scalable-oversightalignmentdebatereward-modelingcapability-gap
可擴展監督的挑戰
隨模型能力增強,如何維持人類監督的技術挑戰。
frontier-researchscalable-oversightalignmentchallenges