A dataset supporting the DICE-RL framework for refining pretrained generative robot policies via reinforcement learning. The data was uploaded by author 'wintermelontree' to Hugging Face on March 20, 2026. It originates from the research paper 'From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning'.
Use Cases
- Finetuning pretrained robot policies based on the described reinforcement learning framework
- Benchmarking robot skill mastery algorithms based on the distribution contractive approach
- Training generative models for robotic control based on the pretraining data mentioned
Strengths
- Dataset is directly linked to a published research paper, providing academic context
- Last updated on 2026-03-20, indicating recent availability
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, file formats, and data size are unknown, which may limit suitability assessment
Provenance
- Source
- wintermelontree on Hugging Face, associated with the research paper 'From Prior to Pro: Efficient Skill Mastery via Distribution Contractive RL Finetuning'
- Collection Method
- Likely generated or collected for the DICE-RL research project
- Time Range
- null
- Freshness
- Last updated 2026-03-20 06:30:43
- Geography
- null