List your company

STEM Reasoning on GPQA Diamond (AUROC)

82.1AUROCSCVC
65.66869.93474.278.466Mar 19, 2026

Evaluation results

MethodLinks
SCVCSampling budget (K)=82026.0382.1
SCVCSampling budget (K)=52026.0381.6
VCSampling budget (K)=82026.0380.8
SCVCSampling budget (K)=22026.0380.3
VCSampling budget (K)=52026.0380.2
VCSampling budget (K)=22026.0376.8
SCSampling budget (K)=82026.0375.4
VCSampling budget (K)=12026.0374.6
SCSampling budget (K)=52026.0373.3
SCSampling budget (K)=22026.0366.3