Robomimicv030: Robot Manipulation Datasets for Latent Reinforcement Learning

A collection of datasets for training and evaluating robot manipulation policies using the ZPRL framework. The data was created by author ManUtdMoon and is associated with a paper titled 'Beyond Action Residuals: Real-World Robot Policy Steering via Bottleneck Latent Reinforcement Learning'. It was last updated on the Hugging Face platform on 2026-05-20.

Use Cases

• Training robot manipulation policies based on the ZPRL framework mentioned in the description
• Evaluating reinforcement learning algorithms for robotics based on the described datasets
• Benchmarking robot policy steering methods based on the bottleneck latent reinforcement learning approach

Strengths

• Associated with a published research paper and a GitHub repository, suggesting academic rigor
• Last updated on 2026-05-20, indicating recent maintenance

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download
• Column-level documentation is absent; field semantics must be inferred after download
• Row count is unknown, which may limit suitability assessment

Provenance

Source

ManUtdMoon

Freshness

Last updated 2026-05-20 15:23:50