Hugging Face's logo

Datasets:
ByteDance
/
MTVQA

Tasks:
Visual Question Answering
Image-to-Text
Modalities:
Image
Text
Formats:
parquet
Languages:
multilingual
Arabic
German
Size:
1K - 10K
ArXiv:
Tags:
multilingual
text-centric
vqa
Libraries:
Datasets
Dask
License:
Dataset card Viewer Files Community
1
MTVQA / default /train
  • 1 contributor
 History: 1 commit
parquet-converter's picture
parquet-converter
Update parquet files
3c22249 verified