Multi-Task Query Understanding on offline Golden (test)
79.89Overall ScoreQP-OneModel
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| QP-OneModelSize=8B2026.02 | 79.89 | 83.86 | 89.86 | 66.17 | 79.68 | |
| QP-OneModelSize=0.6B2026.02 | 78.81 | 81.64 | 89.65 | 65.45 | 78.5 | |
| BERT-PipelineType=Baseline2026.02 | 72.54 | 74.85 | 85.13 | 56.86 | 73.31 |