Hi, I'm Niels from the community science team at Hugging Face!
This pull request improves the dataset card for Omanic by:
- Adding the `question-answering` task category to the metadata.
- Linking the dataset to its original paper: "Omanic: Towards Step-wise Evaluation of Multi-hop Reasoning in Large Language Models".
- Adding a link to the official GitHub repository.
- Providing a brief summary of the dataset splits (OmanicSynth and OmanicBench).
