Datasets:
Modalities:
Text
Formats:
parquet
Languages:
English
Size:
1K - 10K
ArXiv:
Tags:
supervised-fine-tuning
reinforcement-learning
sokoban
general-points
chain-of-thought
instruction-following
Improve dataset card: Add paper/code links, task categories, and usage
#2
by
nielsr
HF Staff
- opened
This PR significantly enhances the dataset card by:
- Adding
task_categories: ['TEXT_GENERATION'],language: ['en'], and relevanttagsto the metadata for better discoverability. - Providing a clear description of the dataset and its context, based on the associated paper.
- Including direct links to the paper (https://huggingface.co/papers/2510.00237) and the GitHub repository (https://github.com/XiaofengLin7/debunking-sft-generalization).
- Incorporating installation instructions and a detailed "Dataset Overview" table from the GitHub README to help users understand the dataset's variants.
- Adding a "Sample Usage" section with code snippets for training models with SFT and GRPO, as found in the GitHub README.
- Including a BibTeX citation for proper academic attribution.
These additions make the dataset card more informative and user-friendly for researchers.
Xiaofeng77
changed pull request status to
merged