Question Answering On Yahoocqa
Métriques
MRR
P@1
Résultats
Résultats de performance de divers modèles sur ce benchmark
Nom du modèle | MRR | P@1 | Paper Title | Repository |
|---|---|---|---|---|
| sMIM (1024) | 0.818 | 0.683 | SentenceMIM: A Latent Variable Language Model | |
| sMIM (1024) + | 0.863 | 0.757 | SentenceMIM: A Latent Variable Language Model | |
| AP-BiLSTM | 0.731 | 0.568 | Attentive Pooling Networks | |
| HyperQA | 0.801 | 0.683 | Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering | |
| CNN | 0.632 | 0.413 | Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering | |
| LSTM | 0.669 | 0.465 | Hyperbolic Representation Learning for Fast and Efficient Neural Question Answering | |
| AP-CNN | 0.726 | 0.560 | Attentive Pooling Networks |
0 of 7 row(s) selected.