Datasets:

---

CausalVerse

/

CausalVerse_Image

like 1

This PR significantly enhances the CausalVerse_Image dataset card by:

• Adding the relevant task_categories (image-feature-extraction, object-detection, video-classification), language: en, license: apache-2.0, and additional tags (causal-representation-learning, simulation, robotics, traffic, physics, synthetic) to the metadata for better discoverability.
• Linking directly to the paper on Hugging Face (2510.14049), the project page (https://causal-verse.github.io/), and the Github repository (https://github.com/CausalVerse/CausalVerseBenchmark).
• Incorporating detailed sections from the Github README including an introduction, overview of the dataset, installation instructions, download/conversion guide, and advanced usage examples with PyTorch, making it much easier for users to get started.
• Adding information about the evaluation benchmarks and the correct citation.
  The content now provides a much richer context and practical guidance for users interacting with the dataset.