What metrics are used to reproduce the evaluation in the leaderboards?

[zhimin-z]

Can you list the metrics correspondence to each benchmark one by one? @matflores @wjessup @bradybd @DarthNerdus Are all of them accuracy?

[acebot712]

The usual metrics corresponding to each dataset are used. However, it would be good to mention that, along with the fact if CoT was used or Zero Shot was used to update the table.

[zhimin-z]

The usual metrics corresponding to each dataset are used. However, it would be good to mention that, along with the fact if CoT was used or Zero Shot was used to update the table.

Thanks for your quick reply. What are the usual metrics corresponding to each dataset? I cannot find any related information in this repository.