跳到正文
@EpochAIResearch· @EpochAIResearch · X·· 21 天前AI 评分35
AI 导读

如果我们无法获取足够信息来审查某个基准,我们会将其标记为"信息不足"。我们会尝试与私有基准的创建者合作进行审查,同时将题目/任务保持在公众知识之外。

正文

If we aren’t able to access enough information to review a benchmark, we’ll designate it ‘Not Enough Info.’ We will try to work with the creators of private benchmarks to conduct reviews while keeping the questions/tasks outside of public knowledge.

来源:@EpochAIResearch · x.com