Question Answering On Drop Test
Métriques
F1
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | F1 |
|---|---|
| question-directed-graph-attention-network-for | 88.38 |
| palm-2-technical-report-1 | 85.0 |
| language-models-are-few-shot-learners | 36.5 |
| gpt-4-technical-report-1 | 80.9 |
| neural-symbolic-reader-scalable-integration | 81.71 |
| numnet-machine-reading-comprehension-with | 67.97 |
| orca-2-teaching-small-language-models-how-to | 60.26 |
| gpt-4-technical-report-1 | 64.1 |
| orca-2-teaching-small-language-models-how-to | 57.97 |
| reasoning-like-program-executors-1 | 87.6 |
| drop-a-reading-comprehension-benchmark | 32.7 |
| giving-bert-a-calculator-finding-operations | 81.78 |
| drop-a-reading-comprehension-benchmark | 47.01 |
| injecting-numerical-reasoning-skills-into | 72.4 |
| a-multi-type-multi-span-network-for-reading | 79.88 |
| tag-based-multi-span-extraction-in-reading | 80.7 |