Jingbo Wang (王靖博) | Embodied AI & Humanoid Robotics

Research Signal

About

I build generalizable humanoid intelligence through physics-based simulation, embodied learning, and sim-to-real transfer.

I am a research scientist at Shanghai AI Lab working on embodied intelligence, humanoid robotics, and physics-based simulation. My research focuses on generalization-oriented humanoid skill learning and sim-to-real transfer. I obtained my PhD from The Chinese University of Hong Kong in 2023, advised by Prof. Dahua Lin.

My work has been published in top-tier venues including CVPR, ICML, ICCV, ECCV, SIGGRAPH, NeurIPS, and ICRA, with several oral, spotlight, and highlight presentations. My publications have been cited more than 10,000 times. I am a recipient of the WAIC 2025 Rising Star Award and winner of the COCO 2018 Panoptic Segmentation Challenge.

Humanoid Robotics Embodied AI Physics Simulation Sim-to-Real

10K+Citations

2025WAIC Rising Star

PhDCUHK

Selected Work

Recent Research Topics

Humanoid control, locomotion, athletic skills, and character animation systems.

01

Behavior Foundation Model

Universal whole-body controllers and teleoperation.

Behavior Foundation Model for Humanoid Robots ICRA 2026

Universal Whole Body Controller with User Friendly Interface

PDF Project Page

AdaMimic: Towards Adaptable Humanoid Control via Adaptive Motion Tracking ICRA 2026

Adaptive Motion Tracking for Humanoid Robots

PDF Code Project Page

UniTracker: Learning Universal Whole-Body Motion Tracker for Humanoid Robots RAL 2026

Universal Whole Body Controller with Kinematics Motion Priors

PDF Project Page

CLOT: Closed-Loop Global Motion Tracking for Whole-Body Humanoid Teleoperation Under Review

Universal Whole Body Controller for PND Adam

PDF Code Project Page

02

Agile Skill Learning

Sports-grade agility learned from human demonstrations.

HumanX: Toward Agile and Generalizable Humanoid Interaction Skills from Human Videos Under Review

Imitation Learning on Basketball Skills

PDF Code Project Page

Learning Athletic Humanoid Tennis Skills from Imperfect Human Motion Data Under Review

Automatic Tennis Skills

PDF Code Project Page

SMASH: Mastering Scalable Whole-Body Skills for Humanoid Ping-Pong with Egocentric Vision Under Review

Table Tennis Skills in Egocentric Vision

PDF Project Page

RoboStriker: Hierarchical Decision-Making for Autonomous Humanoid Boxing Under Review

Multi-Agent Boxing Skills

PDF Project Page

Humanoid Goalkeeper: Learning from Position Conditioned Task-Motion Constraints Under Review

Agile Goalkeeping Skills

PDF Code Project Page

03

Locomotion / Loco-Manipulation

Navigation and interaction policies for constrained terrains.

Gallant: Voxel Grid-based Humanoid Locomotion and Local-Navigation Across 3D Constrained Terrains CVPR 2026 · Highlight

Perceptive Locomotion Controller for Diverse Terrains

PDF Code Project Page

PhysHSI: Towards a Real-World Generalizable Humanoid-Scene Interaction System Under Review

Human Like Loco-Manipulation Policy

PDF Project Page

XHUGWBC: Scalable and General Whole-Body Control for Cross-Humanoid Locomotion ICML 2026

Unified Cross-Humanoid Locomotion and Loco-Manipulation Controller

PDF

04

Character Animation

Large-scale motion synthesis and human-scene interaction.

TokenHSI: Unified Synthesis of Physical Human-Scene Interactions through Task Tokenization CVPR 2025 · Oral

Foundation Controller for Human-Scene Interaction

PDF Code Project Page

Go to Zero: Towards Zero-shot Motion Generation with Million-scale Data ICCV 2025 · Spotlight

Scaled Text-to-Motion Generation Model with Million-scale Data

PDF Code Project Page

AMDM: Interactive Character Control with Auto-Regressive Motion Diffusion Models ACM SIGGRAPH 2025 (TOG)

Control Auto-Regressive Motion Diffusion Models with Reinforcement Learning

PDF Code Project Page

CooHOI: Learning Cooperative Human-Object Interaction with Manipulated Object Dynamics Neurips 2024 · Spotlight

Multi-Agent Cooperative Human-Object Interaction

PDF Code Project Page

UniHSI: Unified Human-Scene Interaction via Prompted Chain-of-Contacts ICLR 2024 · Spotlight

Character Learning with LLMs

PDF Code Project Page

Career

Experience

Research roles across embodied AI, simulation, and robotics systems.

05

Research Career

Full-time and internship positions at leading labs.

Full-time

Research Scientist and Team Lead

Shanghai AI Lab

Sep 2023 – Present

Shanghai, China

Intern

Research Scientist Intern

NVIDIA (Omniverse)

Apr 2022 – Jun 2023

Toronto, Canada

RA

Research Assistant

SenseTime Research

Dec 2019 – Mar 2022

Hong Kong

RA

Research Assistant

CUHK

Aug 2019 – Jul 2020

Hong Kong

Intern

Research Scientist Intern

Momenta

Mar 2019 – Aug 2019

Beijing, China

Intern

Research Scientist Intern

Megvii Research

Mar 2017 – Mar 2019

Beijing, China

Training

Education

Academic background in digital humans, intelligence science, and information science.

06

Academic Background

Degrees from top institutions in China and Hong Kong.

PhD

PhD in Digital Human

The Chinese University of Hong Kong

2020 – 2023

Advised by Prof. Dahua Lin

MPhil

MPhil in Intelligence Science

Peking University

2016 – 2019

BSc

BSc in Information Science

Beijing Institute of Technology

2012 – 2016