Research

Paper

TESTING February 27, 2026

OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control

Authors

Yunshen Wang, Shaohang Zhu, Peiyuan Zhi, Yuhan Li, Jiaxin Li, Yong-Lu Li, Yuchen Xiao, Xingxing Wang, Baoxiong Jia, Siyuan Huang

Abstract

High-fidelity motion tracking serves as the ultimate litmus test for generalizable, human-level motor skills. However, current policies often hit a "generality barrier": as motion libraries scale in diversity, tracking fidelity inevitably collapses - especially for real-world deployment of high-dynamic motions. We identify this failure as the result of two compounding factors: the learning bottleneck in scaling multi-motion optimization and the physical executability constraints that arise in real-world actuation. To overcome these challenges, we introduce OmniXtreme, a scalable framework that decouples general motor skill learning from sim-to-real physical skill refinement. Our approach uses a flow-matching policy with high-capacity architectures to scale representation capacity without interference-intensive multi-motion RL optimization, followed by an actuation-aware refinement phase that ensures robust performance on physical hardware. Extensive experiments demonstrate that OmniXtreme maintains high-fidelity tracking across diverse, high-difficulty datasets. On real robots, the unified policy successfully executes multiple extreme motions, effectively breaking the long-standing fidelity-scalability trade-off in high-dynamic humanoid control.

Metadata

arXiv ID: 2602.23843
Provider: ARXIV
Primary Category: cs.RO
Published: 2026-02-27
Fetched: 2026-03-02 06:04

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.23843v1</id>\n    <title>OmniXtreme: Breaking the Generality Barrier in High-Dynamic Humanoid Control</title>\n    <updated>2026-02-27T09:28:06Z</updated>\n    <link href='https://arxiv.org/abs/2602.23843v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.23843v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>High-fidelity motion tracking serves as the ultimate litmus test for generalizable, human-level motor skills. However, current policies often hit a \"generality barrier\": as motion libraries scale in diversity, tracking fidelity inevitably collapses - especially for real-world deployment of high-dynamic motions. We identify this failure as the result of two compounding factors: the learning bottleneck in scaling multi-motion optimization and the physical executability constraints that arise in real-world actuation. To overcome these challenges, we introduce OmniXtreme, a scalable framework that decouples general motor skill learning from sim-to-real physical skill refinement. Our approach uses a flow-matching policy with high-capacity architectures to scale representation capacity without interference-intensive multi-motion RL optimization, followed by an actuation-aware refinement phase that ensures robust performance on physical hardware. Extensive experiments demonstrate that OmniXtreme maintains high-fidelity tracking across diverse, high-difficulty datasets. On real robots, the unified policy successfully executes multiple extreme motions, effectively breaking the long-standing fidelity-scalability trade-off in high-dynamic humanoid control.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.RO'/>\n    <published>2026-02-27T09:28:06Z</published>\n    <arxiv:primary_category term='cs.RO'/>\n    <author>\n      <name>Yunshen Wang</name>\n    </author>\n    <author>\n      <name>Shaohang Zhu</name>\n    </author>\n    <author>\n      <name>Peiyuan Zhi</name>\n    </author>\n    <author>\n      <name>Yuhan Li</name>\n    </author>\n    <author>\n      <name>Jiaxin Li</name>\n    </author>\n    <author>\n      <name>Yong-Lu Li</name>\n    </author>\n    <author>\n      <name>Yuchen Xiao</name>\n    </author>\n    <author>\n      <name>Xingxing Wang</name>\n    </author>\n    <author>\n      <name>Baoxiong Jia</name>\n    </author>\n    <author>\n      <name>Siyuan Huang</name>\n    </author>\n  </entry>"
}