Datasets:

---

CraftJarvis

/

minecraft-motion-action-dataset

like 1

Follow

CraftJarvis 29

This Pull Request significantly enhances the dataset card for minecraft-motion-action-dataset by providing comprehensive documentation and improving its discoverability on the Hugging Face Hub.

Key updates include:

• Paper Link: Links the dataset to its primary research paper, "Training One Model to Master Cross-Level Agentic Actions via Reinforcement Learning".
• Project Page: Adds a link to the OpenHA project homepage.
• Code Link: Provides a direct link to the associated GitHub repository for the OpenHA suite.
• Task Categories: Incorporates robotics and image-text-to-text task categories, aligning with the dataset's domain and multimodal nature.
• Tags: Adds relevant tags such as minecraft, agent, reinforcement-learning, multimodal, and vision-language-model for improved searchability. language: en is also added.
• Sample Usage: Includes detailed "Installation" and "Inference" instructions with code snippets directly from the GitHub README, enabling users to quickly get started with the dataset and the OpenHA framework.
• Contextual Information: Adds an introduction, interaction details, and an overview of other related datasets within the OpenHA suite to provide a richer understanding of the dataset's role.
• Citation: Includes the BibTeX entry for proper attribution.

These changes make the minecraft-motion-action-dataset card more informative, accessible, and compliant with Hugging Face Hub best practices.