4 ms·
Based on how same models rank fluctuates week to week, all I can conclude is that no frontier models is statistically better than the other or it's too task dep
by wordpad 5mo ago
Based on how same models rank fluctuates week to week, all I can conclude is that no frontier models is statistically better than the other or it's too task dependent that the result cannot converge.