List your company

Math & physics reasoning on MATH 500

82.77AccuracyINFUSER
60.337266.161171.98577.8089Jun 8, 2026

Evaluation results

MethodLinks
INFUSERModel Backbone=Qwen3-8B-Base2026.0682.77
R-FewModel Backbone=Qwen3-8B-Base2026.0682.6
AZRModel Backbone=Qwen3-8B-Base2026.0680.95
R-ZeroModel Backbone=Qwen3-8B-Base2026.0680.55
SPICEModel Backbone=Qwen3-8B-Base2026.0679.4
R-FewModel Backbone=Qwen3-4B-Base2026.0678
SPICEModel Backbone=Qwen3-4B-Base2026.0678
R-ZeroModel Backbone=Qwen3-4B-Base2026.0676.85
INFUSERModel Backbone=Qwen3-4B-Base2026.0676.65
BaseModel Backbone=Qwen3-8B-Base2026.0676.05
AZRModel Backbone=Qwen3-4B-Base2026.0673.9
BaseModel Backbone=Qwen3-4B-Base2026.0661.2