Add dataset card and metadata for ILVR/CoMT

#2
by nielsr HF Staff - opened

This PR adds a dataset card for the CoMT (Chain of Multi-modal Thought) dataset used in the ILVR (Interleaved Latent Visual Reasoning) framework.

The dataset card includes:

  • Metadata for task category (image-text-to-text) and license (mit).
  • Links to the paper and GitHub repository.
  • Information about the data format and directory structure as described in the official repository.
  • Citation information.
Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment