Add dataset card and metadata for ThinkOmni

#2
by nielsr HF Staff - opened

This PR adds a dataset card for the ThinkOmni evaluation datasets. It includes:

  • Links to the paper and the official GitHub repository.
  • Relevant task categories (image-text-to-text, video-text-to-text, audio-text-to-text) for omni-modal reasoning.
  • A summary of the included evaluation benchmarks (MMAU, OmniBench, and Daily-Omni).
  • Citation information.
Catalan258 changed pull request status to merged

Sign up or log in to comment