Natural Language Inference On Rcb
Metriken
Accuracy
Average F1
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | Accuracy | Average F1 |
|---|---|---|
| Modell 1 | 0.418 | 0.302 |
| Modell 2 | 0.518 | 0.357 |
| Modell 3 | 0.546 | 0.406 |
| Modell 4 | 0.463 | 0.367 |
| Modell 5 | 0.498 | 0.306 |
| russiansuperglue-a-russian-language | 0.702 | 0.68 |
| Modell 7 | 0.509 | 0.333 |
| Modell 8 | 0.484 | 0.417 |
| Modell 9 | 0.473 | 0.356 |
| Modell 10 | 0.447 | 0.408 |
| Modell 11 | 0.452 | 0.371 |
| Modell 12 | 0.445 | 0.367 |
| mt5-a-massively-multilingual-pre-trained-text | 0.454 | 0.366 |
| Modell 14 | 0.5 | 0.356 |
| Modell 15 | 0.486 | 0.351 |
| Modell 16 | 0.468 | 0.307 |
| unreasonable-effectiveness-of-rule-based | 0.438 | 0.4 |
| unreasonable-effectiveness-of-rule-based | 0.374 | 0.319 |
| Modell 19 | 0.461 | 0.372 |
| Modell 20 | 0.484 | 0.452 |
| russiansuperglue-a-russian-language | 0.441 | 0.301 |
| unreasonable-effectiveness-of-rule-based | 0.484 | 0.217 |