List your company

Mathematical Reasoning on OOD Math Suite (Math500, GSM8K, Olympiad-Bench)

79.6Pass@1 RateBaseline LLM
59.63264.8167075.184Mar 4, 2026

Evaluation results

MethodLinks
Baseline LLMSystem Category=Standalone2026.0379.610,164100
Brier-SLM-VerbSystem Category=Collaborative2026.0378.89,26457.4
COREA *System Category=Collaborative2026.0377.97,97835.5
RLVR-SLM-ProbeSystem Category=Collaborative2026.0375.95,87425.2
SLM-VerbSystem Category=Collaborative2026.0375.77,30034.9
Router+RLVR-SLMSystem Category=Collaborative2026.0374.44,97420.6
Brier-SLMSystem Category=Standalone2026.0372.72,3220
RLVR-SLM-VerbSystem Category=Collaborative2026.0372.62,92714
RLVR-SLM-AvgProbSystem Category=Collaborative2026.0372.42,2064.5
L1-SLM*System Category=Standalone2026.0372.12,5890
RLVR-SLMSystem Category=Standalone2026.0371.81,9280
SLMSystem Category=Standalone2026.0360.42,6220