List your company

Mathematical Reasoning on AMC 23 (Acc, Tok, ReL)

85AccuracyOurs
53.861.97078.1Feb 1, 2026

Evaluation results

MethodLinks
OursBackbone=Qwen2.5-7B2026.02856,665213.2
Ours (Base)Backbone=Qwen2.5-7B2026.0282.56,704242.9
INFTYTHINKBackbone=Qwen2.5-7B2026.0282.58,061254.3
SepLLMBackbone=Qwen2.5-7B2026.02806,511220.2
LightThinkerBackbone=Qwen2.5-7B2026.02806,790214.6
H2OBackbone=Qwen2.5-7B2026.0277.56,843209.7
OursBackbone=Qwen2.5-1.5B2026.0267.56,460213.2
Ours (Base)Backbone=Qwen2.5-1.5B2026.0262.56,392242.9
INFTYTHINKBackbone=Qwen2.5-1.5B2026.0262.57,605263.5
LightThinkerBackbone=Qwen2.5-1.5B2026.0260.56,312214.6
SepLLMBackbone=Qwen2.5-1.5B2026.0257.56,573220.2
H2OBackbone=Qwen2.5-1.5B2026.02556,230209.7