List your company

Mathematical Reasoning on Math500 (Acc@16)

74Acc@16Verify-Init
52.88858.36963.8569.331Feb 7, 2026

Evaluation results

MethodLinks
Verify-InitBackbone=Qwen2.5-7B-Instruct2026.0274
Mixed-TrainBackbone=Qwen2.5-7B-Instruct2026.0273.5
Verify-AlterBackbone=Qwen2.5-7B-Instruct2026.0272.9
GenerateBackbone=Qwen2.5-7B-Instruct2026.0270.9
Verify-AlterBackbone=Qwen2.5-3B-Instruct2026.0264.5
Verify-InitBackbone=Qwen2.5-3B-Instruct2026.0263.1
GenerateBackbone=Qwen2.5-3B-Instruct2026.0259.6
Mixed-TrainBackbone=Qwen2.5-3B-Instruct2026.0259.2
Verify-InitBackbone=Qwen2.5-1.5B-Instruct2026.0254.7
Verify-AlterBackbone=Qwen2.5-1.5B-Instruct2026.0254.2
Mixed-TrainBackbone=Qwen2.5-1.5B-Instruct2026.0253.8
GenerateBackbone=Qwen2.5-1.5B-Instruct2026.0253.7