Hi, first of all, great work! Just want some clarifications: is all the leaderboard result measured with "no get_document" setting? Is there anyway We can see the eval script for GLM-4.6? I am having a hard time reproducing the exact same result..
Hi, first of all, great work!
Just want some clarifications: is all the leaderboard result measured with "no get_document" setting?
Is there anyway We can see the eval script for GLM-4.6? I am having a hard time reproducing the exact same result..