Objective Evaluation Results of The Approach Considering User Persona on Multi Session Chat.
| Model | BLEU (↑) | BERT-S (↑) | Distinct-1/2 (↑) | PPL (↓) | PerP (↑) | PerB (↑) |
|---|---|---|---|---|---|---|
| BlenderBot | 5.86 | 19.21 | 0.59/3.23 | 5.24 | 9% | 32.35/11.97 |
| Ours (w/o user persona) | 6.02 | 20.06 | 0.58/3.30 | 2.91 | 11% | 31.72/11.59 |
| Ours (w/o keyword expansion) | 6.29 | 76.91 | 0.66/3.47 | 2.85 | 19% | 37.82/16.78 |
| Ours (user persona + keyword expansion) | 6.12 | 83.27 | 0.78/4.38 | 2.72 | 46% | 48.30/25.06 |
| Model | BLEU (↑) | BERT-S (↑) | Distinct-1/2 (↑) | PPL (↓) | PerP (↑) | PerB (↑) |
|---|---|---|---|---|---|---|
| BlenderBot | 5.86 | 19.21 | 0.59/3.23 | 5.24 | 9% | 32.35/11.97 |
| Ours (w/o user persona) | 6.02 | 20.06 | 0.58/3.30 | 2.91 | 11% | 31.72/11.59 |
| Ours (w/o keyword expansion) | 76.91 | 0.66/3.47 | 19% | 37.82/16.78 | ||
| Ours (user persona + keyword expansion) | 6.12 | 2.72 |
Sharing content requires targeting cookies to be enabled. Please update your cookie preferences to use this feature.