AI 导读
CaliBench 论文已通过同行评审，发表在 Transactions on Machine Learning Research（TMLR）上。 我们正在 GitHub 上发布完整的 CaliBench 协议，以便新模型可以直接与论文中的结果进行对比。 是时候向现实攀登了！
正文
The CaliBench paper has been peer-reviewed and published in Transactions on Machine Learning Research (TMLR).
We're releasing the full CaliBench protocol on GitHub, so new models can be compared directly against the results in the paper.
Time to hill-climb reality!
来源：@odysseyml · x.com