YUCHEN
XIAO
I am a Research Scientist and Lead of Embodied AI R&D at YuShu Technology Co., Ltd. Previously, I was a Research Scientist at J.P. Morgan AI Research Group. Before that, I received my Ph.D. from Northeastern University, where I was advised by Prof. Christopher Amato and worked in the Lab for Learning and Planning in Robotics (LLPR). Before I started my Ph.D. program, I obtained my master's degree at Columbia University and worked in the Robotic Manipulation and Mobility (ROAM) Lab.
​
My main research interest is in the field of foundation models, robotics, reinforcement learning and multi-robot systems.
NEWS
-
03/27/2026 We open-source UnifoLM-WBT-Dataset, a high-quality real-wrold humanoid whole-body manipulation dataset with high-frequency rolling updates.
-
03/03/2026 Our work, OmniXtreme: a single flow-based policy with strong generalizability over diverse extreme motions, has been released.
-
02/13/2026 Our work, UnifoLM-X1-0 embodied AI model manufactures robots at Unitree's own robot factory.
-
01/29/2026 Our work, UnifoLM-VLA-0 is released, designed for general-purpose humanoid robot manipulation.
-
11/06/2025 Our work, Embodied Avatar: a whole-body teleoperation platform is released, scaling whole-body manipulation data.
-
09/15/2025 Our work, UnifoLM-WMA-0 is released, a world-model–action architecture spanning multiple types of robotic embodiments.
-
09/23/2024 Our work, Asynchronous Multi-Agent Deep Reinforcement Learning under Partial Observability, has bee accepted to IJRR.
WORKING
RESEARCH
A team of robots collaborates to find and deliver correct tools to two humans.
A Fetch robot is tasked with searching for a target object, a blue Lego block, in clutter.
Two teams of robots (blue vs red) compete to capture the opponent's flag.
Asynchronous Actor-Critic for Multi-Agent Reinforcement Learning
Yuchen Xiao, Weihao Tan and Christopher Amato
The Thirty-Sixth Conference on Neural Information Processing Systems (NeurIPS), 2022, acceptance rate 25.6%
IROS Decision Making in Multi-Agent Systems Workshop, 2022, oral talk
AAAI Symposium: Can We Talk? How to Design Multi-Agent Systems in the Absence of Reliable Communications
A Deeper Understanding of State-Based Critics in Multi-Agent Reinforcement Learning
Xueguang Lyu, Andrea Baisero, Yuchen Xiao and Christopher Amato
The Thirty-Sixth AAAI Conference on Artificial Intelligence, 2022, acceptance rate 15%
Multidisciplinary Conference on Reinforcement Learning and Decision Making (RLDM), 2022, spotlight
Local Advantage Actor-Critic for Robust Multi-Agent Deep Reinforcement Learning
Yuchen Xiao, Xueguang Lyu, and Christopher Amato
IEEE The 3rd International Symposium on Multi-Robot and Multi-Agent Systems (MRS), 2021
* Best Paper Award Finalist *
Contrasting Centralized and Decentralized Critics in Multi-Agent Reinforcement Learning
Xueguang Lyu, Yuchen Xiao, Brett Daley, and Christopher Amato
International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2021, acceptance rate 25%
* Best Paper Award Finalist *
Multi-Agent/Robot Deep Reinforcement Learning with Macro-Actions
Yuchen Xiao, Joshua Hoffman, Tian Xia, and Christopher Amato
Thirty-Fourth AAAI Student Abstract and Poster Program, 2020, spotlight
Learning Multi-Robot Decentralized Macro-Action-Based Policies via a Centralized Q-Net
Yuchen Xiao, Joshua Hoffman, Tian Xia, and Christopher Amato
IEEE International Conference on Robotics and Automation (ICRA), 2020
Macro-Action-Based Deep Multi-Agent Reinforcement Learning
Yuchen Xiao, Joshua Hoffman, and Christopher Amato
Conference on Robot Learning (CoRL), 2019, acceptance rate 27.1%
Online Planning for Target Object Search in Clutter under Partial Observability
Yuchen Xiao, Sammie Katt, Andreas ten Pas, Shengjian Chen, and Christopher Amato
IEEE International Conference on Robotics and Automation (ICRA), 2019
Near-Optimal Adversarial Policy Switching for Decentralized Asynchronous Multi-Agent Systems
Nghia Hoang, Yuchen Xiao (co-first), Kavinayan Sivakumar, Christopher Amato, and Jonathan Patrick How
IEEE International Conference on Robotics and Automation (ICRA), 2018
Contact Localization through Spatially Overlapping Piezoresistive Signals
Pedro Piacenza, Yuchen Xiao (co-first), Steve Park, Ioannis Kymissis, and Matei Ciocarlie
IEEE/RSJ Intelligent Robots and Systems (IROS), 2016
On the Feasibility of Wearable Exotendon Networks for Whole-Hand Movement Patterns in Stroke Patients
Sangwoo Park, Lauri Bishop, Tara Post, Yuchen Xiao, Joel Stein, and Matei Ciocarlie
IEEE International Conference on Robotics and Automation (ICRA), 2016
COURSE PROJECTS
Office Room Service Robot Delivering Object
Northeastern University, Boston, 2017
Robust Grasping for Individual and Cooperative Table Cleaning
Northeastern University, Boston, 2017
Teleoperate PR2 Robot Using Kinect Sensor
Columbia University, New York City, 2015
Baxter Robot Motion Planning Using
RRT&PRM Algorithm
Columbia University, New York City, 2014
[Video]
ACTIVITIES
Senior Program Committee Member, International Joint Conference on Artificial Intelligence (IJCAI), 2021
Program Committee Member, International Conference on Autonomous Agents and Multi-Agent Systems (AAMAS), 2021-2022
Reviewer, IEEE International Conference on Robotics and Automation (ICRA), 2018-2022
Reviewer, Conference on Neural Information Processing Systems (NeurIPS), 2022
Reviewer, IEEE Robotics and Automation Letters (RA-L), 2022
Reviewer, IEEE Transactions on Robotics, 2022
Reviewer, IEEE International Symposium on Multi-Robot and Multi-Agent Systems (MRS), 2021
Reviewer, IEEE International Conference on Intelligent Robots and Systems (IROS), 2021
Reviewer, International Conference on Automated Planning and Scheduling (ICAPS), 2019
Reviewer, Workshop on Reinforcement Learning under Partial Observability (RLPO), NeuralPS 2018
Reviewer, IEEE International Conference on Intelligent Robots and Systems (IROS), 2018
INVITED TALKS
Unitree's Latest Advances and Future Perspectives, IROS, 2025
Macro-Action-Based Multi-Agent Deep Reinforcement Learning, Stanford Intelligent Systems Laboratory, 2020
AWARDS
RAS Travel Grant award for ICRA 2018 in Brisbane, Australia