Epoch continues to do some of the best public benchmarking work on AI. This is helpful (and tells us how bad the state of benchmarking is, and how terrible some of our favorite benchmarks are)
Epoch AI推出Benchmark Reviews项目,揭示当前AI基准测试存在严重缺陷,15个基准中9个被标记为有缺陷。
入选理由:Epoch AI审计发现15个AI基准中9个存在严重缺陷
![[AINews] FrontierCode: Benchmarking for Code Quality over Slop](https://substackcdn.com/image/fetch/$s_!sdBk!,w_1456,c_limit,f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fsubstack-post-media.s3.amazonaws.com%2Fpublic%2Fimages%2F0acd2026-8f85-4504-a5f3-6a0cd82d0b6a_2170x1604.png)





