List your company

Mathematical Reasoning on GSM8K (Accuracy, TFLOPs)

91.43AccuracyOurs-context-unaware
-3.657221.028945.71570.4011Jun 11, 2026

Evaluation results

MethodLinks
Ours-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.0691.439.42
Ours-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.0691.366.56
Ours-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.0684.1513.28
Ours-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.0682.498.9
Ours-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.0682.2617.15
Ours-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.0681.5815.85
T2T-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.0656.7960.83
T2T-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.0651.6337.86
T2T-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.0630.8672.06
T2T-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.0627.9832.75
T2T-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.0618.853.8
T2T-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.0618.5741.95
C2C-context-unawarePair=4B → 8B, Context Setting=context-unaware2026.061.914.94
C2C-context-unawarePair=14B → 4B, Context Setting=context-unaware2026.060.6817.93
C2C-context-unawarePair=8B → 4B, Context Setting=context-unaware2026.060.3814.76
C2C-context-unawarePair=4B → 14B, Context Setting=context-unaware2026.06020.05
C2C-context-unawarePair=8B → 14B, Context Setting=context-unaware2026.0609.84
C2C-context-unawarePair=14B → 8B, Context Setting=context-unaware2026.06024.77