Spaces:
Running
on
CPU Upgrade
Running
on
CPU Upgrade
MATH results have changed
#1102
by
sometimesanotion
- opened
I'm seeing a huge change in the MATH listings for some models, but not all. Are old results being re-worked, or are models being re-evaluated?
They are using a new method which is better and they are reevaluating all the models it seems .. so there will be lot of changes in math scores.
https://huggingface.co/blog/math_verify_leaderboard
Hi @sometimesanotion ,
Yes, @pulkitmehtawork is right, there is a new method to calculate MATH score, so we updated all the models on the Leaderboard
Please, ping me here in case of any questions!
alozowski
changed discussion status to
closed