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 ↗
Do the Rewards Justify the Means? Measuring Trade-Offs Between Rewards and Ethical Behavior in the MACHIAVELLI Benchmark通过带行为标注的交互文本场景,研究奖励追求和伦理权衡。核对原文署名 ↗