SOTAVerified

Multimodal Machine Translation

Multimodal machine translation is the task of doing machine translation with multiple data sources - for example, translating "a bird is flying over water" + an image of a bird over water to German text.

( Image credit: Findings of the Third Shared Task on Multimodal Machine Translation )

Papers

Showing 110 of 108 papers

TitleStatusHype
Seamless: Multilingual Expressive and Streaming Speech TranslationCode6
Attention Is All You NeedCode3
M3P: Learning Universal Representations via Multitask Multilingual Multimodal Pre-trainingCode1
CLIPTrans: Transferring Visual Knowledge with Pre-trained Models for Multimodal Machine TranslationCode1
Distill the Image to Nowhere: Inversion Knowledge Distillation for Multimodal Machine TranslationCode1
Dynamic Context-guided Capsule Network for Multimodal Machine TranslationCode1
3AM: An Ambiguity-Aware Multi-Modal Machine Translation DatasetCode1
BERTGEN: Multi-task Generation through BERTCode1
BigVideo: A Large-scale Video Subtitle Translation Dataset for Multimodal Machine TranslationCode1
Cross-lingual Visual Pre-training for Multimodal Machine TranslationCode1
Show:102550
← PrevPage 1 of 11Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1delMeteor (EN-FR)74.6Unverified
2ERNIE-UniX2BLEU (EN-DE)49.3Unverified
3IKD-MMTBLEU (EN-DE)41.28Unverified
4DCCNBLEU (EN-DE)39.7Unverified
5CaglayanBLEU (EN-DE)39.4Unverified
6Gumbel-Attention MMTBLEU (EN-DE)39.2Unverified
7Multimodal TransformerBLEU (EN-DE)38.7Unverified
8ImagiTBLEU (EN-DE)38.4Unverified
9del+objBLEU (EN-DE)38Unverified
10VMMTFBLEU (EN-DE)37.6Unverified
#ModelMetricClaimedVerifiedStatus
1ViTABLEU (EN-HI)51.6Unverified
#ModelMetricClaimedVerifiedStatus
1ViTABLEU (EN-HI)44.6Unverified