Nexus-VLA: Bridging Human and Robot Experiences via a Shared Hierarchical Semantic Latent Space
Nexus-VLA: Bridging Human and Robot Experiences via a Shared Hierarchical Semantic Latent Space
This research introduces Nexus-VLA, a hybrid policy learning framework that enables cross-embodiment knowledge transfer through end-to-end human-robot joint training. By establishing a hierarchical semantic alignment paradigm from task intent to fine-grained interactions and designing a shared manipulation latent representation, the method allows human manipulation experience to directly contribute to robot policy learning. Experiments across RoboCasa-GR1, out-of-distribution (OOD) settings, and real-world robot experiments demonstrate an 80.5% task success rate on RoboCasa-GR1, along with advantages in human-data utilization, learning efficiency, and cross-scenario generalization.
The paper proposes Nexus-VLA, a hybrid framework for cross-embodiment knowledge transfer using hierarchical semantic alignment and shared latent representations; reported results include an 80.5% task success rate on RoboCasa-GR1 under OOD settings and real-world experiments.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
What the source reports
Publisher-reported claims, with original evidence. These results have not been independently verified by RoboSignal.
- Data origin
- Human egocentric dataOpen source S2
- paper: not_reported
Source excerpts and review record
Automatically extracted; no manual editorial approval recorded.
arXiv:2610.42021v1 Announce Type: new Abstract: Large-scale egocentric human videos provide rich and scalable manipulation experience for robot policy learning, yet substantial embodiment gaps in visual appearance, kinematic structure, and action spaces hinder effective knowledge transfer. Existing methods often decouple cross-embodiment knowledge alignment from robot policy learning, making it difficult for human ma
Open source S2
icy learning efficiency while effectively leveraging human knowledge. Across RoboCasa-GR1, OOD settings, and real-world robot experiments, Nexus-VLA achieves an 80.5% task success rate on RoboCasa-GR1 and demonstrates strong advantages in human-data utilization, learning efficiency, and cross-scenario generalization.
Open source S5
Implications for data suppliers
RoboSignal interpretation and collection questions, not statements of buyer demand.
- Check whether the intended buyer needs human observations, hand or object annotations, or aligned robot actions. Human video alone does not provide robot commands.
- Validate demand and acceptance criteria with a buyer before scaling. Publication, popularity and a research result do not establish a purchase commitment.
Source：arXiv Robotics — research abstracts · arxiv.org