Hugging Face's logo

Datasets:
anvo25
/
vmmu

Tasks:
Image-Text-to-Text
Modalities:
Image
Text
Formats:
parquet
Languages:
Vietnamese
English
Size:
1K - 10K
ArXiv:
Tags:
multimodal
vietnamese
exam
education
vlm
benchmark
Libraries:
Datasets
Dask
Polars
License:
Dataset card Data Studio Files
xet
Community
3

Improve dataset card: Add task category, language, tags, and full abstract

#2
by nielsr HF Staff - opened
base: refs/heads/main
←
from: refs/pr/2
Discussion Files changed
+186
-1
nielsr

This PR enhances the ViExam dataset card by:

  • Adding task_categories: image-text-to-text for better discoverability.
  • Specifying language: vi and en to reflect the dataset's primary language and support for English prompts.
  • Adding relevant tags such as multimodal, vietnamese, exam, education, vlm, benchmark, question-answering, and low-resource.
  • Updating the Abstract section with the full paper abstract.
  • Incorporating "Quick Start Guide", "Getting Started", "Tasks", "Quickstart", "Human-in-the-loop Enhancement", "Repository Structure", and "Key Findings" sections from the GitHub repository to provide comprehensive usage instructions and project details.
 Improve dataset card: Add task category, language, tags, and full abstractcddb9d74
anvo25 changed pull request status to merged
anvo25
 Owner

Hi Niels,
Thank you so much! This is really helpful for us.

Sign up or log in to comment