Command Palette
Search for a command to run...
Natural Language Understanding On Dialoglue
평가 지표
Average
Banking77 (Acc)
CLINC150 (Acc)
DSTC8 (F-1)
HWU64 (Acc)
MultiWOZ (Joint Goal Acc)
Restaurant8k (F-1)
TOP (EM)
평가 결과
이 벤치마크에서 각 모델의 성능 결과
| Paper Title | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| ConvBERT + Pre + Multi | 86.89 | 93.44 | 92.38 | 91.2 | 97.11 | 56.56 | 95.44 | 82.08 | - |
| mslm | 85.83 | 91.17 | 95.8 | 88.33 | 91.36 | 58.22 | 94.85 | 81.1 | - |
| ConvBERT-DG + Pre + Multi | 85.34 | 92.99 | 91.82 | 86.49 | 97.11 | 58.29 | 94.34 | 76.36 | - |
0 of 3 row(s) selected.