Hugging Face
Models
Datasets
Spaces
Posts
Docs
Enterprise
Pricing
Log In
Sign Up
Datasets:
AAAIBenchmark
/
Multi-Opthalingua
like
0
Tasks:
Question Answering
Modalities:
Text
Formats:
text
Languages:
Hindi
Chinese
English
+ 4
Size:
< 1K
Tags:
medical
Libraries:
Datasets
Croissant
License:
mit
Dataset card
Viewer
Files
Files and versions
Community
1
main
Multi-Opthalingua
/
LLMs_Language_bias-main
/
src
/
__pycache__
/
data_analysis.cpython-39.pyc
AAAIBenchmark
Upload 96 files
10bf19f
verified
5 months ago
download
Copy download link
history
contribute
delete
Safe
10.9 kB
This file contains binary data. It cannot be displayed, but you can still
download
it.