Dialogue Generation on Proposed Multi-scenario Dataset 1.0 (test)
86.37Acc TSGM
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| SGMModel=GPT-4o2024.07 | 86.37 | 91.19 | 60.42 | 46.29 | |
| CoT+SOPModel=GPT-4o2024.07 | 82.09 | 86.73 | 57.14 | 43.39 | |
| SGMModel=Llama3-70b2024.07 | 78.35 | 82.86 | 49.18 | 31.87 | |
| SGMModel=Qwen1.5-72b2024.07 | 77.83 | 83.54 | 47.32 | 29.32 | |
| SGMModel=GPT-3.52024.07 | 76.52 | 80.42 | 33.1 | 38.45 | |
| CoT+SOPModel=Qwen1.5-72b2024.07 | 74.25 | 79.65 | 45.24 | 30.72 | |
| CoTModel=GPT-4o2024.07 | 69.64 | 74.34 | 51.3 | 22.72 | |
| CoTModel=Qwen1.5-72b2024.07 | 68.57 | 73.89 | 46.3 | 12.57 | |
| CoTModel=Llama3-70b2024.07 | 65.43 | 72.52 | 44.22 | 11.71 | |
| CoT+SOPModel=GPT-3.52024.07 | 63.24 | 69.49 | 22.22 | 28.57 | |
| SGMModel=Qwen1.5-14b2024.07 | 61.76 | 67.11 | 28.75 | 3.33 | |
| CoT+SOPModel=Qwen1.5-14b2024.07 | 55.64 | 61.57 | 25.67 | 2.69 | |
| CoT+SOPModel=Llama3-70b2024.07 | 52.24 | 54.42 | 40.48 | 12.24 | |
| CoTModel=Qwen1.5-14b2024.07 | 50.47 | 55.75 | 25.93 | 2.15 | |
| CoTModel=GPT-3.52024.07 | 48.57 | 55.75 | 18.52 | 4.83 | |
| SGMModel=Llama3-8b2024.07 | 46.85 | 56.3 | 22.31 | 3.76 | |
| CoT+SOPModel=Llama3-8b2024.07 | 38.72 | 43.28 | 17.14 | 2.26 | |
| CoTModel=Llama3-8b2024.07 | 31.79 | 36.73 | 11.11 | 0.32 |