GALAXY

Reported on 6 benchmarks across 1 task · 1 paper · 5 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Speech6 results

DialogueonMULTIWOZ 2.0
MultiWOZ (Inform)· 2021-11-29
94.4
SOTA
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592
DialogueonMULTIWOZ 2.0
MultiWOZ (Success)· 2021-11-29
85.3
SOTA
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592
DialogueonMULTIWOZ 2.1
BLEU· 2021-11-29
20.01
SOTA
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592
DialogueonMULTIWOZ 2.1
MultiWOZ (Inform)· 2021-11-29
95.3
SOTA
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592
DialogueonMULTIWOZ 2.1
MultiWOZ (Success)· 2021-11-29
86.2
SOTA
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592
DialogueonMULTIWOZ 2.0
BLEU· 2021-11-29
20.5
best: 20.6 (Noisy Channel Model)
GALAXY: A Generative Pre-trained Model for Task-Oriented Dialog with Semi-Supervised Learning and Explicit Policy Injection arXiv:2111.14592