Mathematical Reasoning on Minerva Math (Accuracy, Absolute Improvement, Relative Gain)
22.8AccuracyTTSV
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| TTSVModel=Qwen2.5-Math-7B2025.12 | 22.8 | 9.9 | 76.74 | |
| TTSVModel=LLaMA-3.1-8B2025.12 | 17.6 | 0.3 | 1.73 | |
| LLaMA-3.1-8BModel=LLaMA-3.1-8B2025.12 | 17.3 | — | — | |
| Qwen2.5-Math-7BModel=Qwen2.5-Math-7B2025.12 | 12.9 | — | — | |
| EMModel=Qwen2.5-Math-7B2025.12 | — | 7.8 | — | |
| EMModel=LLaMA-3.1-8B2025.12 | — | 1.9 | — |