Datasets:

---

Seanie-lee

/

ThinkSafe-Qwen3-0.6B-WildGuard

like 0

This PR improves the dataset card for THINKSAFE: Self-Generated Safety Alignment for Reasoning Models.

Changes:

• Added task_categories: text-generation and text-classification
• Added arxiv tag to link the dataset to the paper
• Added dataset description with context and purpose
• Added links to paper and GitHub repository
• Added citation information
• Added relevant tags (safety, alignment, reasoning)

This ensures the dataset can be found when people filter by task_categories and is properly linked to the paper page.

Please review and merge if everything looks good.