Research

Paper

AI LLM March 24, 2026

SIMART: Decomposing Monolithic Meshes into Sim-ready Articulated Assets via MLLM

Authors

Chuanrui Zhang, Minghan Qin, Yuang Wang, Baifeng Xie, Hang Li, Ziwei Wang

Abstract

High-quality articulated 3D assets are indispensable for embodied AI and physical simulation, yet 3D generation still focuses on static meshes, leaving a gap in "sim-ready" interactive objects. Most recent articulated object creation methods rely on multi-stage pipelines that accumulate errors across decoupled modules. Alternatively, unified MLLMs offer a single-stage path to joint static asset understanding and sim-ready asset generation. However dense voxel-based 3D tokenization yields long 3D token sequences and high memory overhead, limiting scalability to complex articulated objects. To address this, we propose SIMART, a unified MLLM framework that jointly performs part-level decomposition and kinematic prediction. By introducing a Sparse 3D VQ-VAE, SIMART reduces token counts by 70% vs. dense voxel tokens, enabling high-fidelity multi-part assemblies. SIMART achieves state-of-the-art performance on PartNet-Mobility and in-the-wild AIGC datasets, and enables physics-based robotic simulation.

Metadata

arXiv ID: 2603.23386
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-24
Fetched: 2026-03-25 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.23386v1</id>\n    <title>SIMART: Decomposing Monolithic Meshes into Sim-ready Articulated Assets via MLLM</title>\n    <updated>2026-03-24T16:16:52Z</updated>\n    <link href='https://arxiv.org/abs/2603.23386v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.23386v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>High-quality articulated 3D assets are indispensable for embodied AI and physical simulation, yet 3D generation still focuses on static meshes, leaving a gap in \"sim-ready\" interactive objects. Most recent articulated object creation methods rely on multi-stage pipelines that accumulate errors across decoupled modules. Alternatively, unified MLLMs offer a single-stage path to joint static asset understanding and sim-ready asset generation. However dense voxel-based 3D tokenization yields long 3D token sequences and high memory overhead, limiting scalability to complex articulated objects. To address this, we propose SIMART, a unified MLLM framework that jointly performs part-level decomposition and kinematic prediction. By introducing a Sparse 3D VQ-VAE, SIMART reduces token counts by 70% vs. dense voxel tokens, enabling high-fidelity multi-part assemblies. SIMART achieves state-of-the-art performance on PartNet-Mobility and in-the-wild AIGC datasets, and enables physics-based robotic simulation.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.GR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.RO'/>\n    <published>2026-03-24T16:16:52Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Chuanrui Zhang</name>\n    </author>\n    <author>\n      <name>Minghan Qin</name>\n    </author>\n    <author>\n      <name>Yuang Wang</name>\n    </author>\n    <author>\n      <name>Baifeng Xie</name>\n    </author>\n    <author>\n      <name>Hang Li</name>\n    </author>\n    <author>\n      <name>Ziwei Wang</name>\n    </author>\n  </entry>"
}