List your company

Reasoning Fidelity on AIME 2025 (ref6)

0.917Top-1 AccuracyCASK
0.85460.87080.8870.9032Apr 13, 2026

Evaluation results

MethodLinks
CASKBudget=5122026.040.9170.9960.25417.50.372
CASKBudget=3842026.040.9030.9910.31318.20.633
TriAttentionBudget=5122026.040.8990.990.32117.50.448
TriAttentionBudget=3842026.040.8910.9880.35718.20.595
CASKBudget=2562026.040.8680.9790.50417.50.559
TriAttentionBudget=2562026.040.8570.9780.517.50.634