Dialogue Evaluation on USR-PersonaChat (test)
0.495Pearson Correlation (r)USL-H
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| USL-Htrained_on_dataset=false2023.09 | 0.495 | 0.523 | |
| RADEtrained_on_dataset=false2023.09 | 0.451 | 0.465 | |
| USRtrained_on_dataset=false2023.09 | 0.44 | 0.418 | |
| GRADEtrained_on_dataset=false2023.09 | 0.358 | 0.352 | |
| METEORtrained_on_dataset=false2023.09 | 0.253 | 0.271 | |
| BERTScoretrained_on_dataset=false2023.09 | 0.152 | 0.122 | |
| BLEURTtrained_on_dataset=false2023.09 | 0.065 | 0.054 |