List your company

Mathematical Reasoning on Numina

45.5AccuracyAcquisitionSynthesis (Confidence)
16.69224.17131.6539.129May 13, 2026

Evaluation results

MethodLinks
AcquisitionSynthesis (Confidence)Model=Llama 8B, Category=AS2026.0545.5
FilteredModel=Llama 8B, Category=Selection2026.0545.2
AcquisitionSynthesis (Diversity)Model=Qwen 7B, Category=AS2026.0545.2
Prismatic Syn.Model=Qwen 3B, Category=Synthesis2026.0544.6
OriginalModel=Llama 8B, Category=Synthesis2026.0544.5
RandomModel=Llama 8B, Category=Selection2026.0543.7
AcquisitionSynthesis (Gradient Norm)Model=Llama 8B, Category=AS2026.0543.7
AcquisitionSynthesis (Answer Variance)Model=Qwen 7B, Category=AS2026.0543.4
AcquisitionSynthesis (Answer Variance)Model=Llama 8B, Category=AS2026.0543.2
AcquisitionSynthesis (Format)Model=Qwen 7B, Category=AS2026.0543.2
AcquisitionSynthesis (Format)Model=Llama 8B, Category=AS2026.0543.1
AcquisitionSynthesis (Confidence)Model=Qwen 7B, Category=AS2026.0543.1
AcquisitionSynthesis (Diversity)Model=Llama 8B, Category=AS2026.0542.8
AcquisitionSynthesis (Gradient Norm)Model=Qwen 7B, Category=AS2026.0542.5
AcquisitionSynthesis (Proximity)Model=Llama 8B, Category=AS2026.0540.7
AcquisitionSynthesis (Proximity)Model=Qwen 7B, Category=AS2026.0540.2
OriginalModel=Qwen 7B, Category=Synthesis2026.0536.5
FilteredModel=Qwen 7B, Category=Selection2026.0534.8
Prismatic Syn.Model=Qwen 7B, Category=Synthesis2026.0534.8
AcquisitionSynthesis (Confidence)Model=Qwen 3B, Category=AS2026.0533.7
Prismatic Syn.Model=Llama 8B, Category=Synthesis2026.0533
DataEnvGymModel=Qwen 3B, Category=Synthesis2026.0532.3
Untrained modelModel=Llama 8B2026.0532.2
AcquisitionSynthesis (Answer Variance)Model=Qwen 3B, Category=AS2026.0530.5
AcquisitionSynthesis (Gradient Norm)Model=Qwen 3B, Category=AS2026.0529.7
AcquisitionSynthesis (Format)Model=Qwen 3B, Category=AS2026.0529.6
OriginalModel=Qwen 3B, Category=Synthesis2026.0529.4
Untrained modelModel=Qwen 7B2026.0529.3
FilteredModel=Qwen 3B, Category=Selection2026.0529.2
RandomModel=Qwen 3B, Category=Selection2026.0528.8
Untrained modelModel=Qwen 3B2026.0528.3
DataEnvGymModel=Llama 8B, Category=Synthesis2026.0526.8
AcquisitionSynthesis (Diversity)Model=Qwen 3B, Category=AS2026.0524.7
RandomModel=Qwen 7B, Category=Selection2026.0522.9
AcquisitionSynthesis (Proximity)Model=Qwen 3B, Category=AS2026.0522.1
DataEnvGymModel=Qwen 7B, Category=Synthesis2026.0517.8