Dialogue Policy Evaluation on PersonaChat (test)
97.7USR RETFIXED
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| FIXED2022.02 | 97.7 | — | 0.852 | — | 81.3 | 0.933 | 0.25 | |
| Humans2022.02 | 92.7 | 1.623 | 0.267 | 7.512 | 0.24 | 0.951 | 0.153 | |
| PARROT2022.02 | 92.5 | — | 0.794 | 8.933 | 0.01 | 0.898 | 0.223 | |
| LC2022.02 | 87.9 | 1.572 | 0.103 | 6.769 | 0.11 | 0.918 | 0.195 | |
| BL2022.02 | 84.7 | 1.507 | 0.185 | 6.797 | 0.06 | 0.844 | 0.07 | |
| HF2022.02 | 79.4 | 1.48 | 0.272 | 6.707 | 0.23 | 0.925 | 0.152 | |
| BR2022.02 | 72.5 | 1.35 | 0.211 | 6.12 | 0.2 | 0.946 | 0.138 | |
| KV2022.02 | 53.8 | 1.147 | 0.217 | 4.982 | 0.23 | 0.852 | 0.122 | |
| DR2022.02 | 48.2 | 0.906 | 0.268 | 4.779 | 0.02 | 0.537 | 0.22 | |
| S22022.02 | 39 | 0.681 | 0.121 | 3.814 | 0.13 | 0.845 | 0.111 |