Hugging Face
Models
Datasets
Spaces
Posts
Docs
Solutions
Pricing
Log In
Sign Up
Datasets:
ByteDance
/
MTVQA
like
10
Tasks:
Visual Question Answering
Image-to-Text
Modalities:
Image
Text
Formats:
parquet
Languages:
multilingual
Arabic
German
+ 7
Size:
1K - 10K
ArXiv:
arxiv:
2405.11985
Tags:
multilingual
text-centric
vqa
Libraries:
Datasets
Dask
Croissant
+ 1
License:
cc-by-nc-4.0
Dataset card
Viewer
Files
Files and versions
Community
1
refs/convert/parquet
MTVQA
/
default
/
test
/
0000.parquet
Commit History
Update parquet files
3c22249
verified
parquet-converter
commited on
May 24