XMimic and XSoma: Learning Humanoid Motion Tracking and Locomotion in Under a Minute

We present XMimic and XSoma, which learn humanoid motion tracking and locomotion on rough terrain in under a minute of training. Both follow SuperMimic, which learns a world model of the robot and optimizes the policy with analytic gradients through short rollouts imagined in this model. Existing implementations of this method do not make use of larger GPUs, as every training iteration is a long chain of small GPU operations that run one after another. XMimic and XSoma therefore shorten this chain and run the steps of every iteration at the same time on up to four GPUs. We compile the complete training loop with JAX, including the MuJoCo Warp simulator. For locomotion, XSoma leaves only the dynamics of the robot to the world model and computes everything else exactly, including the terrain height under the feet. It learns with a horizon of one second and without a critic. On four H100 GPUs, XMimic learns to track all 40 LAFAN1 clips with the Unitree G1 in 0.86 minutes, 3.7 times faster than BlinkMimic on one GPU. XSoma reaches the tracking target that Holosoma's tests set for FastSAC in 0.77 minutes, 12.3 times faster than FastSAC on one GPU, while needing only an eighth of its environment steps.

Authors

Publication Details

Journal
arXiv (Cornell University)
Published
2026-11-16
DOI
https://doi.org/10.48550/arxiv.2611.20452
Primary Topic
Reinforcement Learning in Robotics
Type
preprint
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
preprint

XMimic and XSoma: Learning Humanoid Motion Tracking and Locomotion in Under a Minute

Nico Bohlinger, Jan Peters
arXiv (Cornell University)
Reinforcement Learning in Robotics
preprint

XMimic and XSoma: Learning Humanoid Motion Tracking and Locomotion in Under a Minute

Nico Bohlinger, Jan Peters
preprint en

Abstract

We present XMimic and XSoma, which learn humanoid motion tracking and locomotion on rough terrain in under a minute of training. Both follow SuperMimic, which learns a world model of the robot and optimizes the policy with analytic gradients through short rollouts imagined in this model. Existing implementations of this method do not make use of larger GPUs, as every training iteration is a long chain of small GPU operations that run one after another. XMimic and XSoma therefore shorten this chain and run the steps of every iteration at the same time on up to four GPUs. We compile the complete training loop with JAX, including the MuJoCo Warp simulator. For locomotion, XSoma leaves only the dynamics of the robot to the world model and computes everything else exactly, including the terrain height under the feet. It learns with a horizon of one second and without a critic. On four H100 GPUs, XMimic learns to track all 40 LAFAN1 clips with the Unitree G1 in 0.86 minutes, 3.7 times faster than BlinkMimic on one GPU. XSoma reaches the tracking target that Holosoma's tests set for FastSAC in 0.77 minutes, 12.3 times faster than FastSAC on one GPU, while needing only an eighth of its environment steps.

arXiv (Cornell University)
Reinforcement Learning in Robotics
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon