arx5_real_world_datasets

This repository contains the real-world robot datasets collected on the ARX-5 platform as presented in the paper Self-Correcting VLA: Online Action Refinement via Sparse World Imagination.

The dataset is part of the SC-VLA framework, which enhances robotic physical grounding through intrinsic self-improvement using Sparse World Imagination (SPI) and Online Action Refinement (OAR).

• Code: https://github.com/Kisaragi0/SC-VLA
• Paper: Self-Correcting VLA: Online Action Refinement via Sparse World Imagination

Usage

You can download the dataset using the HuggingFace CLI:

huggingface-cli download Kisaragi0/arx5_real_world_datasets \
  --repo-type dataset \
  --local-dir arx5_real_world_datasets

Citation

@article{SC-VLA,
      title={Self-Correcting VLA: Online Action Refinement via Sparse World Imagination}, 
      author={Chenyv Liu and Wentao Tan and Lei Zhu and Fengling Li and Jingjing Li and Guoli Yang and Heng Tao Shen},
      journal={arXiv preprint arXiv:2602.21633},
      year={2026},
}