Authority Intent Classification on Authority Intent (test)
77.34AccuracyQP-OneModel-8B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| QP-OneModel-8BSize=8B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 77.34 | 72.76 | |
| Qwen3-32BSize=32B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 77.14 | 72.02 | |
| Qwen3-8BSize=8B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 76.8 | 71.94 | |
| RedOne-7BSize=7B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 76.76 | 71.74 | |
| Qwen2.5-7B-InstructSize=7B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 76.46 | 71.41 | |
| RedOne2.0-8BSize=8B, Evaluation Protocol=Supervised Fine-Tuning (SFT)2026.02 | 76.36 | 71.11 | |
| Qwen3-32BSize=32B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 71.62 | 64.16 | |
| QP-OneModel-8BSize=8B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 71.56 | 60.41 | |
| RedOne2.0-8BSize=8B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 70.56 | 63 | |
| Qwen3-8BSize=8B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 59.97 | 49.48 | |
| RedOne-7BSize=7B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 48.78 | 39.96 | |
| Qwen2.5-7B-InstructSize=7B, Evaluation Protocol=In-Context Learning (Few-shot)2026.02 | 34.5 | 28.1 |