3260 papers • 126 benchmarks • 313 datasets
Translate a given word in a source language to a word in the target language, given the source sentence and one or more images illustrating the word.
(Image credit: Papersgraph)
These leaderboards are used to track progress in multimodal-machine-translation
Use these libraries to find multimodal-machine-translation models and implementations
No subtasks available.
A large-scale multimodal and multilingual dataset that aims to facilitate research on grounding words to images in their contextual usage in language is introduced and a fill-in-the-blank task is proposed to demonstrate the utility of the dataset.
Adding a benchmark result helps the community track progress.