General Reasoning on TheoremQA (accuracy)
32.47Accuracy (General Reasoning)Qwen2.5-7B-Instruct
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-7B-InstructSetting=Teacher2026.05 | 32.47 | |
| Hard KDSetting=Student (Qwen2.5-1.5B)2026.05 | 19.23 | |
| HybKDSetting=Student (Qwen2.5-1.5B)2026.05 | 18.65 | |
| Skew Reverse KLSetting=Student (Qwen2.5-1.5B)2026.05 | 17.62 | |
| Forward KL divergenceSetting=Student (Qwen2.5-1.5B)2026.05 | 17.15 | |
| Total VariationSetting=Student (Qwen2.5-1.5B)2026.05 | 17.08 | |
| Reverse KL divergenceSetting=Student (Qwen2.5-1.5B)2026.05 | 17 | |
| Skew Forward KLSetting=Student (Qwen2.5-1.5B)2026.05 | 16.25 | |
| Baseline (No Distill)Setting=Student (Qwen2.5-1.5B)2026.05 | 9.9 |