Question Answering On Fquad 1
评估指标
EM
F1
评测结果
各个模型在此基准测试上的表现结果
| Paper Title | Repository | |||
|---|---|---|---|---|
| CamemBERT-Large | 82.1 | 92.2 | FQuAD: French Question Answering Dataset | - |
| XLM-RoBERTa-Large | 79.0 | 89.5 | FQuAD: French Question Answering Dataset | - |
| CamemBERT-Base | 78.4 | 88.4 | FQuAD: French Question Answering Dataset | - |
| Camembert-Base-SquadFR-Fquad-Piaf | 77.0 | 88.4 | Project PIAF: Building a Native French Question-Answering Dataset | |
| XLM-RoBERTa-Base | 75.3 | 85.9 | FQuAD: French Question Answering Dataset | - |
| Fmikaelian-Camembert-Base-Fquad | 75.0 | 85.6 | - | - |
| LePetit | 57.2 | 70.71 | On the importance of pre-training data volume for compact language models | - |
0 of 7 row(s) selected.