Mathematical Reasoning on MATH 500 (Accuracy and Micro Average)
69.4AccuracyTraceLift
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| TraceLiftModel=Qwen2.5-14B2026.05 | 69.4 | 72.25 | |
| Exec-onlyModel=Qwen2.5-14B2026.05 | 68.6 | 71.93 | |
| BaseModel=Qwen2.5-14B2026.05 | 67 | 71.2 | |
| TraceLiftModel=Qwen2.5-7B2026.05 | 55.2 | 70.19 | |
| Exec-onlyModel=Qwen2.5-7B2026.05 | 54 | 65.62 | |
| BaseModel=Qwen2.5-7B2026.05 | 42.4 | 46.51 | |
| TraceLiftModel=Llama3.1-8B2026.05 | 15.4 | 22.86 | |
| Exec-onlyModel=Llama3.1-8B2026.05 | 14.8 | 22.19 | |
| BaseModel=Llama3.1-8B2026.05 | 10.2 | 12.74 |