Dialogue Generation On Fusedchat
평가 지표
BLEU
Inform
Inform_mct
Joint SA
PPL
SSA
Sensibleness
Slot Accuracy
Specificity
Success
Success_mct
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | BLEU | Inform | Inform_mct | Joint SA | PPL | SSA | Sensibleness | Slot Accuracy | Specificity | Success | Success_mct | Paper Title | Repository |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Two-in-one model | 12.05 | 70.4 | 90.1 | 0.592 | 10.49 | 0.50 | 0.52 | 0.972 | 0.47 | 57.0 | 72.7 | Fusing task-oriented and open-domain dialogues in conversational agents | |
| Classification-based model | 12.17 | 75.1 | 90.8 | 0.600 | 10.50 | 0.55 | 0.58 | 0.973 | 0.51 | 60.9 | 74.4 | Fusing task-oriented and open-domain dialogues in conversational agents |
0 of 2 row(s) selected.