Dialogue Evaluation on ConvAI2
0.554Pearson CorrelationQuantiDCE
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| QuantiDCE2021.06 | 0.554 | 0.554 | 0.395 | 0.501 | |
| GRADEtraining_dataset=DailyDialog++2021.06 | 0.496 | 0.503 | 0.356 | 0.452 | |
| BERT-RUBERtraining_dataset=DailyDialog++2021.06 | 0.266 | 0.266 | 0.185 | 0.239 | |
| BERTScore2021.06 | 0.225 | 0.225 | 0.154 | 0.201 | |
| BLEURTfine-tuned=true2021.06 | 0.152 | 0.149 | 0.103 | 0.135 | |
| METEOR2021.06 | 0.145 | 0.181 | 0.123 | 0.15 | |
| ROUGE2021.06 | 0.136 | 0.14 | 0.097 | 0.124 | |
| ADEMtraining_dataset=DailyDialog++2021.06 | 0.026 | 0.037 | 0.049 | 0.037 | |
| BLEU2021.06 | 0.003 | 0.128 | 0.088 | 0.073 |