GitHub Blog · AI & ML · Michelle Zhou · 28 天前精选模型评分71
GitHub 发布 人工智能（AI，Artificial Intelligence） 代码评审开放基准 ReviewBench
ReviewBench: An open benchmark for AI code review
导读
GitHub 发布 人工智能（AI，Artificial Intelligence） 代码评审开放基准 ReviewBench，包含 219 个公开拉取请求、覆盖 19 种语言，并公开数据集、评分规则和评审模型。
推荐理由
GitHub 公开了基准数据集、评分规则和评审模型，读者可以据此复现评测，并了解离线结果与线上实验的对应关系。
分类置信度
高
- 官方或准官方来源，且无冲突类别。
- 正文、可信日期、主分类和证据摘录完整。
来源级别：官方一手
事件类型：研究发布
证据摘录
- “We built ReviewBench , a new code review offline benchmark , to address that gap, and it is available for you to use today.”
- “ReviewBench’s research preview version is now available through the ReviewBench website , where you can explore the full benchmark, compare code review agents, and bring your own agent to evaluate and iterate.”
- “Run against a 25-PR test set with per-PR detail and repeat as you tune your configuration.”
- “When you’re ready, run the full set of 219 pull requests (three rounds), scored by the same judge as every other entry.”
- “Your scores remain private until a maintainer reviews and approves the submission.”
来源：GitHub Blog · AI & ML · github.blog