Task-oriented Dialogue on FewShotSGD unseen schemata (test)
28.76BLEUFull Data Baseline
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Full Data BaselineTrain split=164,9782021.10 | 28.76 | 1.54 | |
| BLEURT self-trainingPseudo-response selection strategy=BLEURT, Train split=40-shot (4,312)2021.10 | 27.53 | 2.72 | |
| BLEURT self-trainingPseudo-response selection strategy=BLEURT, Train split=20-shot (2,140)2021.10 | 27.38 | 3.77 | |
| Vanilla self-trainingPseudo-response selection strategy=Vanilla, Train split=40-shot (4,312)2021.10 | 26.61 | 4.2 | |
| T5-smallPseudo-response selection strategy=None, Train split=40-shot (4,312)2021.10 | 26.52 | 5.97 | |
| BLEURT self-trainingPseudo-response selection strategy=BLEURT, Train split=10-shot (1,075)2021.10 | 25.49 | 3.82 | |
| Vanilla self-trainingPseudo-response selection strategy=Vanilla, Train split=20-shot (2,140)2021.10 | 25.47 | 9.11 | |
| T5-smallPseudo-response selection strategy=None, Train split=20-shot (2,140)2021.10 | 25.14 | 11.51 | |
| Vanilla self-trainingPseudo-response selection strategy=Vanilla, Train split=10-shot (1,075)2021.10 | 24.38 | 7.67 | |
| BLEURT self-trainingPseudo-response selection strategy=BLEURT, Train split=5-shot (558)2021.10 | 24.13 | 5.39 | |
| T5-smallPseudo-response selection strategy=None, Train split=10-shot (1,075)2021.10 | 22.79 | 14.98 | |
| Vanilla self-trainingPseudo-response selection strategy=Vanilla, Train split=5-shot (558)2021.10 | 21.97 | 15.96 | |
| T5-smallPseudo-response selection strategy=None, Train split=5-shot (558)2021.10 | 20.52 | 19.93 |