HEX: Humanoid-Aligned Experts for Cross-Embodiment Whole-Body Manipulation

arXiv:2604.07993v1 Announce Type: new Abstract: Humans achieve complex manipulation through coordinated whole-body control, whereas most Vision-Language-Action (VLA) models treat robot body parts largely independently, making high-DoF humanoid control challenging and often unstable. We present HEX, a state-centric framework for coordinated manipulation on full-sized bipedal humanoid robots. HEX introduces a humanoid-aligned universal state representation for scalable learning across heterogeneous embodiments, and incorporates a Mixture-of-Experts Unified Proprioceptive Predictor to model whole-body coordination and temporal motion dynamics from large-scale multi-embodiment trajectory data. To efficiently capture temporal visual context, HEX uses lightweight history tokens to summarize past observations, avoiding repeated encoding of historical images during inference. It further employs a residual-gated fusion mechanism with a flow-matching action head to adaptively integrate visual-language cues with proprioceptive dynamics for action generation. Experiments on real-world humanoid manipulation tasks show that HEX achieves state-of-the-art performance in task success rate and generalization, particularly in fast-reaction and long-horizon scenarios.

Related

• HiF-VLA: Hindsight, Insight and Foresight through Motion Representation for Vision-Language-Action Models
• UniLACT: Depth-Aware RGB Latent Action Learning for Vision-Language-Action Models
• ViVa: A Video-Generative Value Model for Robot Reinforcement Learning
• Drift-Based Policy Optimization: Native One-Step Policy Learning for Online Robot Control
• 'Don't Do That!': Guiding Embodied Systems through Large Language Model-based Constraint Generation

Source: arXiv cs.RO | 2026-04-10