MorphMimic: Morphology-Conditioned World Model and Policy Sharing Across Humanoids
Lizhi Yang, Fengze Xie*, Yuyang Wang*, Lu Gan, Yisong Yue, Aaron D. Ames
MorphMimic is a cross-embodiment world model and whole-body motion tracker built on SuperMimic, capable of controlling the Unitree G1, AgiBot X2 Ultra, and Booster T2 humanoid robots.

PhysCoRe: Physics-Corrected Residual World Models for Material-Aware Deformable Dynamics
Haocheng Yin*, Shuohan Tao*, Yongsheng Chen, Lu Gan
CoRL 2026
PhysCoRe is a physics-aware residual world model for deformable object dynamics that generalizes to unseen objects and enables uncertainty-guided online active exploration.

ShelfGaussian: Shelf-Supervised Open-Vocabulary Gaussian-based 3D Scene Understanding
Lingjun Zhao*, Yandong Luo*, James Hays, Lu Gan
CVPR 2026 Findings
ShelfGaussian is an open-vocabulary multi-modal Gaussian-based 3D scene understanding framework supervised by off-the-shelf vision foundation models.

• Paper
• Website
• Code

HERCULES: An Open-Source Simulation Framework for Heterogeneous Multi-Robot SLAM, Collaborative Perception, and Exploration
Sandilya Sai Garimella, Daniel Chase Butterfield, Sean Wilson, Lu Gan
CVPR 2026 Workshop on Multi-Agent Robotic Systems
HERCULES is a UE5-based multi-robot simulator with out-of-the-box support for heterogeneous multi-robot SLAM, collaborative perception, and closed-loop coordination.

Beyond Topology: A Morphological Symmetry Graph Representation for Locomotion Policy Learning
Sizhe Wei*, Xulin Chen*, Fengze Xie, Garrett Ethan Katz, Zhenyu Gan, Lu Gan
MS-PPO is a PPO variant featuring a morphological-symmetry-equivariant actor and a morphological-symmetry-invariant critic network.

• Paper
• Website

WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems
Yuchen Wang*, Jiangtao Kong*, Sizhe Wei, Xiaochang Li, Haohong Lin, Hongjue Zhao, Tianyi Zhou, Lu Gan, Huajie Shao
ICML 2026, Spotlight (Top 2.2%)
WestWorld is a cross-embodied trajectory world model for robot control, pretrained on 89 environments spanning diverse robot morphologies across simulation and real-world data.

• Paper
• Website
• Code

GaussianFormer3D: Multi-Modal Gaussian-based Semantic Occupancy Prediction with 3D Deformable Attention
Lingjun Zhao, Sizhe Wei, James Hays, Lu Gan
ICRA 2026, RSS 2025 Gaussian Representations for Robot Autonomy Workshop
GaussianFormer3D is a LiDAR-camera semantic occupancy prediction framework based on a compact and continuous object-centric scene representation.

• Paper
• Website
• Code
• Video

STATE-NAV: Stability-Aware Traversability Estimation for Bipedal Navigation on Rough Terrain
Ziwon Yoon, Lawrence Y. Zhu, Jingxi Lu, Lu Gan, Ye Zhao
IEEE RA-L, 2025
STATE-NAV is a learning-based humanoid traversability estimation framework for generalized rough terrain navigation by incorporating bipedal locomotion stability.

• Paper
• Website
• Code
• Video

Morphological-Symmetry-Equivariant Heterogeneous Graph Neural Network for Robotic Dynamics Learning
Fengze Xie*, Sizhe Wei*, Yue Song, Yisong Yue, Lu Gan
L4DC, 2025
MS-HGNN is an efficient deep learning model for robotic dynamics learning, that integrates both robotic kinematic structures and morphological symmetries.

• Paper
• Website
• Poster
• Code

MI-HGNN: Morphology-Informed Heterogeneous Graph Neural Network for Legged Robot Contact Perception
Daniel Butterfield, Sandilya Sai Garimella, Nai-Jen Cheng, Lu Gan
ICRA, 2025
MI-HGNN is a structured deep learning model that incorporates robot morphological constraints; designed as a learning-based contact estimator for legged robots.

• Paper
• Website
• Poster
• Code
• Video