Xiaolong Wang

Learning Massively Multitask World Models for Continuous Control

3D Aware Region Prompted Vision Language Model

TIPS: Turn-level Information-Potential Reward Shaping for Search-Augmented LLMs

XDex: Learning Cross-Embodiment Dexterous Grasping with 1000 Hands

House Of Dextra : Cross-Embodied Co-Design for Dexterous Hands

In-and-On: Scaling Egocentric Manipulation with in-the-wild and on-task Data

Leveraging Physics-Based Models for Rapid Adaptation in Reinforcement Learning

EgoVLA: Learning Vision-Language-Action Models from Egocentric Human Videos

Real Deep Research from Foundation Model to Robotics

Learning to (Learn at Test Time): RNNs with Expressive Hidden States

Dynamic Gaussians Mesh: Consistent Mesh Reconstruction from Dynamic Scenes

Consistent Flow Distillation for Text-to-3D Generation

Hierarchical World Models as Visual Whole-Body Humanoid Controllers

3D-SPATIAL MULTIMODAL MEMORY

Gaussian-Augmented Physics Simulation and System Identification with Complex Colliders

HandsOnVLM: Vision-Language Models for Hand-Object Interaction Prediction

Learning to (Learn at Test Time): RNNs with Expressive Hidden States

NaVILA: Legged Robot Vision-Language-Action Model for Navigation

Learning Dynamic 3D Gaussians from Monocular Videos without Camera Poses