Skip to search formSkip to main contentSkip to account menu

Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization

This work introduces Being-H0.5, a foundational Vision-Language-Action model designed for robust cross-embodiment generalization across diverse robotic platforms, and designs a unified sequential modeling and multi-task pre-training paradigm to bridge human demonstrations and robotic execution.
arXiv (opens in a new tab)

Rethinking Visual-Language-Action Model Scaling: Alignment, Mixture, and Regularization

A systematic, controlled study of VLA scaling that revisits core training choices for pretraining across diverse robots and introduces a Grouped Blind Ensemble protocol that blinds operators to model identity and separates policy execution from outcome judgment, reducing experimenter bias.
arXiv (opens in a new tab)

Joint-Aligned Latent Action: Towards Scalable VLA Pretraining in the Wild

JALA is presented, a pretraining framework that learns Jointly-Aligned Latent Actions, a transition-aware, behavior-centric latent space for learning from heterogeneous human data that generates more realistic hand motions in both controlled and unconstrained scenarios.
arXiv (opens in a new tab)

Conservative Offline Robot Policy Learning via Posterior-Transition Reweighting

PTR reallocates credit according to how attributable each sample's post-action consequence is under the current representation, improving conservative offline adaptation to heterogeneous robot data.
arXiv (opens in a new tab)
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)