Introducing SWE-bench Verified

公司动态 OpenAI 原文语言:英文

We’re releasing a human-validated subset of SWE-bench that more reliably evaluates AI models’ ability to solve real-world software issues.

原文摘录

We’re releasing a human-validated subset of SWE-bench that more reliably evaluates AI models’ ability to solve real-world software issues.

以下是来源提供的简短摘录,完整内容请查看原文。

参考来源

← 返回每日简报