Siyuan Huang

DiffWind: Physics-Informed Differentiable Modeling of Wind-Driven Object Dynamics

Flow Connecting Actions and Reactions: A Condition-Free Framework for Human Action-Reaction Synthesis

SceneCOT: Eliciting Grounded Chain-of-Thought Reasoning in 3D Scenes

G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior

VideoArtGS: Building Digital Twins of Articulated Objects from Monocular Video

GaussianFluent: Gaussian Simulation for Dynamic Scenes with Mixed Materials

GapONet: Nonlinear Operator Learning for Bridging the Humanoid Sim-to-Real Gap

LEO-VL: Efficient Scene Representation for Scalable 3D Vision-Language Learning

Taccel: Scaling Up Vision-based Tactile Robotics via High-performance GPU Simulation

SceneWeaver: All-in-One 3D Scene Synthesis with an Extensible and Self-Reflective Agent

Building Interactable Replicas of Complex Articulated Objects via Gaussian Splatting

Task-oriented Sequential Grounding in 3D Scenes

MOVIS: Enhancing Multi-Object Novel View Synthesis for Indoor Scenes