PeopleofSuperintelligence

Researcher

Yifeng Jiang

Yifeng Jiang is a Research Scientist at NVIDIA whose work spans robot control and human-motion simulation. His degrees include a bachelor's degree from Shanghai Jiao Tong University, an MS in electrical and computer engineering from Georgia Tech, and a computer science PhD from Stanford, advised by C. Karen Liu. He coauthored SimGAN on adapting learned policies, Transformer Inertial Poser on real-time motion reconstruction from six IMUs, and Nymeria, a multimodal motion dataset collected in the wild. In 2025, he coauthored MaskedManipulator and ProtoMotions3; in July 2026, he coauthored GPC, which learns reusable motor skills from motion data.

Now at
NVIDIA

Timeline

  1. 2018

    Learning human joint limits

    In 2018, he coauthors an ICRA paper with Liu that learns realistic human joint limits from real human data.

  2. 2019

    Human motion with joint torques

    In 2019, he coauthors a SIGGRAPH paper that converts a muscle-actuation optimal-control problem into an equivalent joint-torque formulation. The method generates human-like movement with simpler modeling and faster computation.

  3. 2021

    SimGAN for robot transfer

    In 2021, he coauthors SimGAN, which identifies a hybrid physics simulator by matching simulated trajectories to a target domain. The resulting simulator helps adapt learned policies to target dynamics.

  4. 2022

    Motion from six IMUs

    In 2022, he coauthors Transformer Inertial Poser, which reconstructs full-body motion in real time from six wearable IMUs and generates the terrain a person has visited.

  5. 2024

    Nymeria and HMD²

    In 2024, he coauthors Nymeria, a large multimodal dataset of human motion collected in the wild with egocentric devices. He also coauthors HMD², which generates full-body motion from a single head-mounted device with an outward-facing camera and visual SLAM.

  6. 2025

    MaskedManipulator and ProtoMotions3

    In 2025, he coauthors MaskedManipulator, a generative control policy for whole-body manipulation that accepts high-level object or body pose goals. His 2025 work also includes ProtoMotions3, an open-source GPU-accelerated framework for simulated humanoids.

  7. July 2026

    Generative motor control

    In July 2026, he coauthors GPC, which uses end-to-end reinforcement learning to learn reusable motor skills from large-scale motion data. Its generative controller can be adapted for different tasks.

  8. As of October 2026

    Research scientist at NVIDIA

    As of October 2026, he is a Research Scientist at NVIDIA. His degrees include a bachelor's degree from Shanghai Jiao Tong University, an MS in electrical and computer engineering from Georgia Tech, and a computer science PhD from Stanford, advised by C. Karen Liu.

Career

  1. NowNVIDIAResearch Scientist

Education

  1. Shanghai Jiao Tong UniversityBachelor's degree
  2. Georgia Institute of TechnologyMS
  3. Stanford UniversityPhD in Computer Science

From the biography.

  1. Google has the highest combined h-index among AI organizationsIncludes NVIDIA. We estimate that Google holds the highest combined h-index at 8,016, while universities account for 73% of the deduplicated total across AI organizations.
  2. 20% of people at companies who shape AI also hold a university role, led by Google and Stanford UniversityIncludes NVIDIA. We estimate that 539 of the 6,766 people with a known current organization (8%) hold both a university role and a company role, including 20% of people at companies and 13% of people at universities.
  3. Google alumni have founded or led 133 other organizations, more than any other AI labIncludes NVIDIA. We estimate that alumni of 17 major AI labs went on to found or lead 350 other organizations, including 226 they founded, and we trace 472 career moves directly between 23 labs since 2010.
  4. Universities (31%) and startups and other companies (25%) employ the largest shares of 2020–2026 entrantsIncludes NVIDIA. Among 819 people whose first recorded role started in 2020–2026 and who have a known current organization, 60% work in commercial sectors (25% at startups and other companies, 23% at frontier labs and 13% at big tech, with 9 in two), and 31% work at universities, compared with 66% for earlier entrants.

Sources and credits

Photo: We have no published photo of Yifeng Jiang yet. Send us a free-licensed photo.

  1. Data-Driven Approach to Simulating Realistic Human Joint Constraints arxiv.org. Checked 21 October 2026.
  2. Synthesis of Biologically Realistic Human Motion Using Joint Torque Actuation arxiv.org. Checked 21 October 2026.
  3. SimGAN: Hybrid Simulator Identification for Domain Adaptation via Adversarial Reinforcement Learning arxiv.org. Checked 21 October 2026.
  4. Transformer Inertial Poser: Real-time Human Motion Reconstruction from Sparse IMUs with Simultaneous Terrain Generation arxiv.org. Checked 21 October 2026.
  5. Nymeria: A Massive Collection of Multimodal Egocentric Daily Motion in the Wild arxiv.org. Checked 21 October 2026.
  6. HMD²: Environment-aware Motion Generation from Single Egocentric Head-Mounted Device arxiv.org. Checked 21 October 2026.
  7. MaskedManipulator: Versatile Whole-Body Manipulation arxiv.org. Checked 21 October 2026.
  8. Yifeng Jiang - Stanford University cs.stanford.edu. Checked 21 October 2026.
  9. Presentation - SIGGRAPH 2026 Conference Schedule s2026.conference-schedule.org. Checked 21 October 2026.
  10. Yifeng Jiang | DAIR @ NVIDIA Research research.nvidia.com. Checked 21 October 2026.