SOP Prediction on Proposed SOP dataset
80.25PrecisionSFT
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| SFTModel=Qwen1.5-72b2024.07 | 80.25 | 74.01 | 77 | 2.86 | 7.12 | |
| DALModel=GPT-4o2024.07 | 78.11 | 66.51 | 71.85 | 2.01 | 5.4 | |
| SFTModel=Llama3-70b2024.07 | 74.1 | 73.03 | 73.56 | 2.81 | 7.27 | |
| SFTModel=Llama3-8b2024.07 | 72.19 | 68.33 | 70.21 | 3.54 | 9.33 | |
| TCoTModel=GPT-4o2024.07 | 69.2 | 73.34 | 71.22 | 4.7 | 12.46 | |
| SFTModel=Qwen1.5-14b2024.07 | 68.58 | 62.73 | 65.52 | 3.94 | 10.3 | |
| DALModel=Llama3-70b2024.07 | 64.14 | 56.05 | 59.82 | 3.85 | 10.72 | |
| TCoTModel=Llama3-70b2024.07 | 60.81 | 67.34 | 63.91 | 5.43 | 14.38 | |
| TCoTModel=GPT-3.52024.07 | 50.95 | 50.03 | 50.48 | 6.91 | 18.28 | |
| TCoTModel=Qwen1.5-72b2024.07 | 48.29 | 51.94 | 50.04 | 7.38 | 18.72 | |
| DALModel=Qwen1.5-72b2024.07 | 46.74 | 31.57 | 37.68 | 10.75 | 28.32 | |
| TCoTModel=Llama3-8b2024.07 | 44.5 | 40.35 | 42.32 | 10.92 | 30.1 | |
| DALModel=Qwen1.5-14b2024.07 | 42.84 | 37.81 | 40.17 | 10.23 | 25.77 | |
| DALModel=GPT-3.52024.07 | 41.39 | 21.94 | 28.68 | 7.61 | 22.48 | |
| TCoTModel=Qwen1.5-14b2024.07 | 38.79 | 42.69 | 40.65 | 11.36 | 26.91 | |
| DALModel=Llama3-8b2024.07 | 35.19 | 28.04 | 31.21 | 8.31 | 24.05 |