Passage Retrieval On Peerqa
Metriken
MRR
Recall@10
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | MRR | Recall@10 |
|---|---|---|
| splade-v3-new-baselines-for-splade | 0.4536 | 0.6851 |
| how-to-train-your-dragon-diverse-augmentation | 0.4845 | 0.6817 |
| Modell 3 | 0.4288 | 0.6388 |
| towards-unsupervised-dense-information | 0.3624 | 0.5567 |
| knowledge-distillation-of-large-language | 0.4839 | 0.6709 |
| large-dual-encoders-are-generalizable | 0.4142 | 0.6122 |
| towards-unsupervised-dense-information | 0.4408 | 0.6314 |
| colbertv2-effective-and-efficient-retrieval | 0.4122 | 0.6371 |