Hao Su

Learning Massively Multitask World Models for Continuous Control

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

Coupled Distributional Random Expert Distillation for World Model Online Imitation Learning

LVSPM: Long Sequence View Synthesis and Pose Estimation Model

AuxVLA: Auxiliary Task Learning and Multi-Modal Enhancement for Vision-Language-Action Models in Mobile Manipulation

Dynamic Gaussians Mesh: Consistent Mesh Reconstruction from Dynamic Scenes

Policy Decorator: Model-Agnostic Online Refinement for Large Policy Model

Staggered Environment Resets Improve Massively Parallel On-Policy Reinforcement Learning

Point-SAM: Promptable 3D Segmentation Model for Point Clouds

Hierarchical World Models as Visual Whole-Body Humanoid Controllers

Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning

MeshLRM: Large Reconstruction Model for High-Quality Meshes

Reward-free World Models for Online Imitation Learning

Reward-free World Models for Online Imitation Learning

ManiSkill-HAB: A Benchmark for Low-Level Manipulation in Home Rearrangement Tasks

3D-Adapter: Geometry-Consistent Multi-View Diffusion for High-Quality 3D Generation

ElastoGen: 4D Generaetive Elastodynamics

TF-HOT: Training-Free Hand-Object Pose Tracking and Optimization for Dexterous Manipulation