Horizon-Labs's picture
Multilingual prompt task & complexity classifier small v1.0
dbd52f5 verified
Raw History Blame Contribute Delete
542 Bytes
{
"Horizon-Labs/multilingual-prompt-task-complexity-small": {
"n": 4200,
"task_acc": 0.7930952380952381,
"task_macro_f1": 0.734950807495426,
"task_acc_lang_mean": 0.7930952380952382,
"task_acc_en": 0.83,
"spearman_creativity": 0.7689230428861341,
"spearman_reasoning": 0.7132914236351086,
"spearman_contextual_knowledge": 0.5912620228625542,
"spearman_domain_knowledge": 0.661298466421794,
"spearman_constraints": 0.8776595387619498,
"spearman_complexity": 0.6814550922131966,
"spearman_mean": 0.7156482644634562
}
}