Compliant Residual DAgger: Improving Real-World Contact-Rich Manipulation with Human Corrections

Xiaomeng Xu*, Yifan Hou*, Chendong Xin, Zeyi Liu, Shuran Song
Neural Information Processing Systems (Neurips 2025)
Human-to-Robot Workshop (CoRL 2025), Best Paper
Website  •   Paper  •   Code

TL;DR: Improving contact-rich manipulation with a compliant residual policy learned from on-policy delta human correction.

Vision in Action: Learning Active Perception from Human Demonstrations

Haoyu Xiong, Xiaomeng Xu, Jimmy Wu, Yifan Hou, Jeannette Bohg, Shuran Song
Conference on Robot Learning (CoRL 2025)
Website  •   Paper  •   Video  •   Hardware&Code

TL;DR: Learning where to look with a robot neck for bimanual manipulation.

RoboPanoptes: The All-Seeing Robot with Whole-body Dexterity

Xiaomeng Xu, Dominik Bauer, Shuran Song
Robotics: Science and Systems (RSS 2025)
Website  •   Paper  •   Video  •   Hardware&Code

TL;DR: A robot that utilizes all of its body parts to sense and interact with the environment.

Dynamics-Guided Diffusion Model for Sensor-less Robot Manipulator Design

Xiaomeng Xu, Huy Ha, Shuran Song
Conference on Robot Learning (CoRL 2024)
MAPoDeL Workshop (CoRL 2024), Best Machine Learning Paper
Assembly Workshop (CoRL 2024), Spotlight Presentation
Website  •   Paper  •   Video  •   Code

TL;DR: Learning to design task-specific manipulators without task-specific training.

JacobiNeRF: NeRF Shaping with Mutual Information Gradients

Xiaomeng Xu, Yanchao Yang, Kaichun Mo, Boxiao Pan, Li Yi, Leonidas J. Guibas
Conference on Computer Vision and Pattern Recognition (CVPR 2023)
Website  •   Paper  •   Video  •   Code

TL;DR: Shaping a NeRF to encode mutual correlations in a scene via aligning Jacobians.

Enhancing Generalizable 6D Pose Tracking of an In-Hand Object with Tactile Sensing

Yun Liu*, Xiaomeng Xu*, Weihang Chen, Haocheng Yuan, He Wang, Jing Xu, Rui Chen, Li Yi
Robotics and Automation Letters (RA-L 2023) with a presentation at ICRA 2024
Paper  •   Video  •   Code

TL;DR: Tactile-enhanced 6D pose tracking of previously unseen in-hand objects.

AutoGPart: Intermediate Supervision Search for Generalizable 3D Part Segmentation

Xueyi Liu, Xiaomeng Xu, Anyi Rao, Chuang Gan, Li Yi
Conference on Computer Vision and Pattern Recognition (CVPR 2022)
Website  •   Paper  •   Video  •   Code

TL;DR: Improving the generalizability of 3D part segmentation networks by searching for intermediate supervisions.

ViviPaint: Creating Dynamic Painting with a Thermochromic Toolkit

Guanhong Liu, Tianyu Yu, Zhihao Yao, Haiqing Xu, Yunyi Zhang, Xuhai Xu, Xiaomeng Xu, Mingyue Gao, Qirui Sun, Tingliang Zhang, Haipeng Mi
Multimodal Technologies and Interaction (MTI 2022)
Website  •   Paper

TL;DR: A toolkit that assists artists and enthusiasts in creating thermochromic paintings.