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.