List your company

Aggregate Reasoning Performance on Overall

93.51AccuracyROM_CSC
84.607686.918889.2391.5412Mar 23, 2026

Evaluation results

MethodLinks
ROM_CSCCounterfactual Self-Correction=true2026.0393.511,15912.37
EAT2026.0393.471,8436.26
L12026.0393.071,3138.1
Cut2048max tokens=20482026.0392.771,8425.79
ROM2026.0392.531,4219.14
Vanilla2026.0391.722,1975.6
Cut1024max tokens=10242026.0391.291,7435.9
Cut512max tokens=5122026.0384.951,6805.83