Improve dataset card: Add audio-text-to-text task category, library, and tags

#2
by nielsr HF Staff - opened

This PR improves the CCFQA dataset card by enhancing its discoverability and utility.
Specifically, it:

  • Adds audio-text-to-text to task_categories to accurately reflect its cross-modal nature for speech and text factuality evaluation.
  • Includes library_name: datasets for direct integration into the Hugging Face datasets library.
  • Adds comprehensive tags such as multilingual, cross-modal, speech, text, factuality, and evaluation to provide richer context and improve searchability.

The paper link remains the existing arXiv link as per guidelines.

yxdu changed pull request status to merged

Sign up or log in to comment