Service Dialogue Task on Service dialogue scenarios
52CRSEAD
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| SEADParams=14B2026.02 | 52 | 9.6 | 0.912 | 0.63 | 1.57 | 1.55 | 0 | |
| GPT-4o2026.02 | 44.2 | 10.8 | 0.867 | 0.04 | 0.97 | 1.34 | 727.28 | |
| LongCat-FlashParams=560B2026.02 | 42.2 | 10 | 0.925 | 0.28 | 1.33 | 1.56 | 23.08 | |
| Qwen2.5-72B-InstructParams=72B2026.02 | 39 | 9.6 | 0.818 | 0.51 | 1.06 | 1.18 | 0 | |
| Qwen2.5-14B-InstructParams=14B2026.02 | 38.7 | 10.5 | 0.883 | 0.34 | 0.68 | 0.63 | 0 | |
| Qwen2.5-32B-InstructParams=32B2026.02 | 38.3 | 9.9 | 0.899 | -0.11 | 0.76 | 2.25 | 0 | |
| Qwen3-235BParams=235B2026.02 | 32.3 | 10.4 | 0.765 | -0.24 | 0.8 | 1.54 | 69.36 | |
| DeepSeek-ChatParams=671B2026.02 | 31.6 | 11.3 | 0.863 | -0.2 | 0.27 | 0.76 | 87.36 |