IKD-MMT

Reported on 6 benchmarks across 2 tasks · 1 paper · 6 SOTA

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

Natural Language Processing6 results

Machine TranslationonMulti30K
BLEU (EN-DE)· 2022-10-10
41.28
best: 49.3 (ERNIE-UniX2)
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468
Machine TranslationonMulti30K
Meteor (EN-DE)· 2022-10-10
58.93
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468
Machine TranslationonMulti30K
Meteor (EN-FR)· 2022-10-10
77.2
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468
Multimodal Machine TranslationonMulti30K
BLEU (EN-DE)· 2022-10-10
41.28
best: 49.3 (ERNIE-UniX2)
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468
Multimodal Machine TranslationonMulti30K
Meteor (EN-DE)· 2022-10-10
58.93
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468
Multimodal Machine TranslationonMulti30K
Meteor (EN-FR)· 2022-10-10
77.2
SOTA
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine Translation arXiv:2210.04468