OpenAI News·· 2024-08-13AI 评分50
OpenAI 发布 SWE-bench Verified
Introducing SWE-bench Verified
AI 导读
OpenAI 推出经过人工验证的 SWE-bench 子集,旨在更可靠地评估 AI 模型解决真实世界软件问题的能力。
原文
We’re releasing a human-validated subset of SWE-bench that more reliably evaluates AI models’ ability to solve real-world software issues.
来源:OpenAI News · openai.com