[hacker news] How We Broke Top AI Agent Benchmarks: And What Comes Next]

摘要

How We Broke Top AI Agent Benchmarks: And What Comes Next]

来源

本文首发于 hacker news。

阅读原文：[How We Broke Top AI Agent Benchmarks: And What Comes Next]](https://rdi.berkeley.edu/blog/trustworthy-benchmarks-cont/)