Query Reformulation on 8,500 Body queries Reformulation V1
97.3Precision@1Gold (n=100)
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| Gold (n=100)Relevance Criteria=Loose Relevance2026.01 | 97.3 | 97.8 | 97.6 | |
| PRECISE-PPI (λ=0.95)Relevance Criteria=Loose Relevance2026.01 | 94.3 | 94.5 | 94.2 | |
| Sonnet-Unlb (binary)Relevance Criteria=Strict Relevance2026.01 | 85.2 | 84.5 | 84 | |
| Sonnet-Unlb (binary)Relevance Criteria=Loose Relevance2026.01 | 85.2 | 84.5 | 84 | |
| Sonnet-Unlb (prob)Relevance Criteria=Strict Relevance2026.01 | 77.4 | 77.4 | 77.4 | |
| Sonnet-Unlb (prob)Relevance Criteria=Loose Relevance2026.01 | 77.4 | 77.4 | 77.4 | |
| Gold (n=100)Relevance Criteria=Strict Relevance2026.01 | 64.6 | 64.1 | 65.7 | |
| PRECISE-PPI (λ=0.95)Relevance Criteria=Strict Relevance2026.01 | 59.5 | 59.1 | 60.3 |