Improve dataset card: Add description, links, sample usage, and experiments

#2
by nielsr HF Staff - opened

This PR significantly improves the KoWit-24 dataset card by:

  • Updating the title to the full paper title.
  • Adding direct links to the Hugging Face paper page, the GitHub repository, the Hugging Face Dataset itself, and the LangChain prompts page.
  • Replacing placeholder sections with detailed content from the GitHub README, including a comprehensive overview, dataset description, key features, and an example dataset entry.
  • Populating the wordplay type distribution table (Table 1).
  • Adding extensive sample usage code snippets for loading the dataset, performing automatic interpretation evaluation, and running experiments with other LLMs, all sourced directly from the GitHub README.
  • Including a dedicated section for "Experiments and Results" with the detailed description of tasks and the results table from the GitHub README.
  • Adding the BibTeX citation for the paper.

These changes provide a much more complete and useful resource for users interested in the KoWit-24 dataset.

Ready to merge
This branch is ready to get merged automatically.

Sign up or log in to comment