Query Understanding and Criteria Generation
94.8Semantic SimilarityWisModel
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| WisModel2025.12 | 94.8 | 74.9 | 54.4 | 67.7 | 39.8 | 98.2 | |
| GPT-4o2025.12 | 91.3 | 64 | 39.4 | 52.6 | 21.5 | 142.2 | |
| DeepSeek-V3.2-Expsource_type=open-source2025.12 | 90.2 | 59.3 | 32.4 | 48 | 14.4 | 153.5 | |
| GPT-52025.12 | 87 | 53.8 | 27.6 | 41.8 | 13.2 | 163.3 | |
| GLM-4.62025.12 | 84.8 | 55.5 | 30.2 | 44.5 | 14.4 | 168.1 | |
| GLM-4-Flash2025.12 | 82.2 | 50 | 25.8 | 42.1 | 9.9 | 167.1 | |
| Qwen-Max2025.12 | 78.1 | 43.2 | 23.1 | 35.8 | 11.8 | 168.9 |