List your company

Mathematical Reasoning on HMMT 2026 (val)

37.1AccuracyiS@1
25.03628.16831.334.432Jun 7, 2026

Evaluation results

MethodLinks
iS@1Sample size (N)=32, Max length=16,000 tokens, Top-k=12026.0637.1
Self-CertaintySample size (N)=32, Max length=16,000 tokens, Selection variant=best over each dataset2026.0634.8
Self-ConsistencySample size (N)=32, Max length=16,000 tokens2026.0628.5
DeepConfSample size (N)=32, Max length=16,000 tokens, Selection variant=best over each dataset2026.0625.5