Dialogue Evaluation On Usr Topicalchat
Métriques
Pearson Correlation
Spearman Correlation
Résultats
Résultats de performance de divers modèles sur ce benchmark
Tableau comparatif
Nom du modèle | Pearson Correlation | Spearman Correlation |
---|---|---|
usr-an-unsupervised-and-reference-free | 0.4220 | 0.4192 |
proxy-indicators-for-the-quality-of-open | 0.4974 | 0.4877 |
mdd-eval-self-training-on-augmented-data-for | 0.4575 | 0.5109 |
usr-an-unsupervised-and-reference-free | 0.4068 | 0.3245 |
usr-an-unsupervised-and-reference-free | 0.3221 | 0.1419 |
usr-an-unsupervised-and-reference-free | 0.3345 | 0.3086 |