Add safeagents judge bench results

#2
by adaamko - opened
vLLM Semantic Router org

our safeagents judge bench v0 run, test split, unmodified benchmark runner, fixed 0.5 cut-off, no tuning. revision 462efc366229

balanced accuracy 0.758, AUROC 0.878, false-block rate 0.436, category macro F1 0.245

Xunzhuo changed pull request status to merged

Sign up or log in to comment