Generative Question Answering On Cicero
평가 지표
ROUGE
평가 결과
이 벤치마크에서 각 모델의 성능 결과
모델 이름 | ROUGE | Paper Title | Repository |
|---|---|---|---|
| T5-large pre-trained on GLUCOSE | 0.2980 | CICERO: A Dataset for Contextualized Commonsense Inference in Dialogues | |
| T5-large | 0.2946 | CICERO: A Dataset for Contextualized Commonsense Inference in Dialogues | |
| T5-large pre-trained on COMET | 0.2878 | CICERO: A Dataset for Contextualized Commonsense Inference in Dialogues | |
| BART | 0.2837 | CICERO: A Dataset for Contextualized Commonsense Inference in Dialogues |
0 of 4 row(s) selected.