activity
20242026
collaborators

8 papers

math.OC2026

Accelerated Mirror Descent Method through Variable and Operator Splitting

Long Chen, Hao Luo, Jingrong Wei +2

Mirror descent uses the mirror function to encode geometry and constraints, improving convergence while preserving feasibility. Accelerated Mirror Descent Methods (Acc-MD) are deri…

cs.RO2026

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

Hao Luo, Ye Wang, Wanpeng Zhang +9

We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs…

cs.RO2025

Spatial-Aware VLA Pretraining through Visual-Physical Alignment from Human Videos

Yicheng Feng, Wanpeng Zhang, Ye Wang +4

Vision-Language-Action (VLA) models provide a promising paradigm for robot learning by integrating visual perception with language-guided policy learning. However, most existing ap…

cs.AI2025

Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds

Weihao Tan, Xiangyang Li, Yunhao Fang +11

We introduce Lumine, the first open recipe for developing generalist agents capable of completing hours-long complex missions in real time within challenging 3D open-world environm…

cs.CV2025

Being-H0: Vision-Language-Action Pretraining from Large-Scale Human Videos

Hao Luo, Yicheng Feng, Wanpeng Zhang +7

We introduce Being-H0, a dexterous Vision-Language-Action model (VLA) trained on large-scale human videos. Existing VLAs struggle with complex manipulation tasks requiring high dex…

cs.CV2025

Unified Multimodal Understanding via Byte-Pair Visual Encoding

Wanpeng Zhang, Yicheng Feng, Hao Luo +4

Multimodal large language models (MLLMs) have made significant progress in vision-language understanding, yet effectively aligning different modalities remains a fundamental challe…