Common Sense Reasoning On Parus
Metriken
Accuracy
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | Accuracy |
|---|---|
| Modell 1 | 0.574 |
| unreasonable-effectiveness-of-rule-based | 0.498 |
| russiansuperglue-a-russian-language | 0.486 |
| Modell 4 | 0.908 |
| Modell 5 | 0.508 |
| Modell 6 | 0.766 |
| Modell 7 | 0.528 |
| unreasonable-effectiveness-of-rule-based | 0.478 |
| Modell 9 | 0.598 |
| Modell 10 | 0.508 |
| Modell 11 | 0.584 |
| mt5-a-massively-multilingual-pre-trained-text | 0.504 |
| unreasonable-effectiveness-of-rule-based | 0.48 |
| Modell 14 | 0.562 |
| russiansuperglue-a-russian-language | 0.982 |
| Modell 16 | 0.492 |
| Modell 17 | 0.66 |
| Modell 18 | 0.498 |
| Modell 19 | 0.498 |
| Modell 20 | 0.476 |
| Modell 21 | 0.676 |
| Modell 22 | 0.554 |