Numerical Reasoning on GSM8K (test)
1.13MAE (Scale 1)EMO
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| EMOBackbone=Qwen2.5-1.5B, Loss Function=EMO2026.05 | 1.13 | 1.08 | 0.85 | 0.98 | 1.72 | 0.71 | |
| NTLBackbone=Qwen2.5-1.5B, Loss Function=NTL2026.05 | 1.16 | 1.06 | 0.84 | 1.05 | 1.53 | 0.78 | |
| DELBackbone=Qwen2.5-1.5B, Loss Function=Digit Entropy Loss2026.05 | 1.17 | 1.03 | 0.81 | 0.91 | 1.17 | 0.63 | |
| DIST2LossBackbone=Qwen2.5-1.5B, Loss Function=DIST2Loss2026.05 | 1.18 | 1.14 | 0.93 | 1.11 | 1.62 | 0.77 | |
| MixCEBackbone=Qwen2.5-1.5B, Loss Function=MixCE2026.05 | 1.25 | 1.18 | 0.91 | 1.08 | 1.72 | 0.79 | |
| MLEBackbone=Qwen2.5-1.5B, Loss Function=MLE2026.05 | 1.26 | 1.2 | 0.91 | 1.11 | 1.57 | 0.72 |