标签
这条推文批判了传统的AI基准测试，并介绍了TRACES，这是一个新的基准测试，专注于评估模型如何得出答案的过程，包括工具使用、错误纠正和证据追踪。
This paper introduces Apodex Discovery, a framework and reality benchmark (TRACES) for evaluating and building 'discoverative AI'—AI systems that investigate open-ended real-world problems. The proposed heavy-duty solvers outperform published state-of-the-art on AAV capsid design and improve drug repurposing prediction scores.