General Reasoning on Average (HumanEval, GSM8K, Math-500, MT-Bench)
4.82SpeedUpPARD-2
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PARD-2Target Model=Qwen3-8B, Inference Draft Length=16, Temperature=02026.05 | 4.82 | 6.14 | |
| PARD-2Target Model=Qwen3-14B, Inference Draft Length=16, Temperature=02026.05 | 4.79 | 6.03 | |
| PARD-2Target Model=Qwen3-32B, Inference Draft Length=16, Temperature=02026.05 | 4.68 | 5.75 | |
| PARDTarget Model=Qwen3-14B, Inference Draft Length=16, Temperature=02026.05 | 4.51 | 5.14 | |
| PARDTarget Model=Qwen3-8B, Inference Draft Length=16, Temperature=02026.05 | 4.39 | 5.56 | |
| PARDTarget Model=Qwen3-32B, Inference Draft Length=16, Temperature=02026.05 | 4.37 | 5.33 | |
| ARTarget Model=Qwen3-8B, Temperature=02026.05 | 1 | 1 | |
| ARTarget Model=Qwen3-14B, Temperature=02026.05 | 1 | 1 | |
| ARTarget Model=Qwen3-32B, Temperature=02026.05 | 1 | 1 |