MLE-bench: Evaluating Machine Learning Agents on Machine Learning Engineering通过竞赛任务和人类表现基线评测自主机器学习工程能力。核对原文署名 ↗
Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety将推理链监测作为安全研究机会,并分析其可靠性可能降低的条件。核对原文署名 ↗