Text Generation On Redial
Métriques
Distinct-2
Distinct-3
Distinct-4
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
| Nom du modèle | Distinct-2 | Distinct-3 | Distinct-4 |
|---|---|---|---|
| c2-crs-coarse-to-fine-contrastive-learning | 0.189 | 0.334 | 0.424 |
| crfr-improving-conversational-recommender | 0.345 | 0.516 | 0.639 |
| improving-conversational-recommender-systems | 0.289 | 0.434 | 0.519 |
| towards-knowledge-based-recommender-dialog | - | 0.3 | 0.45 |
| towards-unified-conversational-recommender | 0.492 | 0.648 | 0.832 |