# CoSMo3D: Open-World Promptable 3D Semantic Part Segmentation through LLM-Guided Canonical Spatial Modeling

arXiv: https://arxiv.org/abs/2603.01205 · HTML: https://mlsanity.com/paper/2603.01205/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 34 references that are in the map.
- First posted on arXiv 1 Mar 2026.

## Builds on (most cited first, 26 of 26)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Segment Anything](https://mlsanity.com/paper/2304.02643/) (2023)
- [Sigmoid Loss for Language Image Pre-Training](https://mlsanity.com/paper/2303.15343/) (2023)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [SAM 2: Segment Anything in Images and Videos](https://mlsanity.com/paper/2408.00714/) (2024)
- [Point Transformer](https://mlsanity.com/paper/2011.00931/) (2020)
- [Point Transformer V3: Simpler, Faster, Stronger](https://mlsanity.com/paper/2312.10035/) (2023)
- [PartNet: A Large-scale Benchmark for Fine-grained and Hierarchical Part-level 3D Object Understanding](https://mlsanity.com/paper/1812.02713/) (2018)
- [Normalized Object Coordinate Space for Category-Level 6D Object Pose and Size Estimation](https://mlsanity.com/paper/1901.02970/) (2019)
- [Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework](https://mlsanity.com/paper/2202.07123/) (2022)
- [OpenMask3D: Open-Vocabulary 3D Instance Segmentation](https://mlsanity.com/paper/2306.13631/) (2023)
- [PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning](https://mlsanity.com/paper/2211.11682/) (2022)
- [Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material](https://mlsanity.com/paper/2506.15442/) (2025)
- [SAM3D: Segment Anything in 3D Scenes](https://mlsanity.com/paper/2306.03908/) (2023)
- [CraftsMan3D: High-fidelity Mesh Generation with 3D Native Generation and Interactive Geometry Refiner](https://mlsanity.com/paper/2405.14979/) (2024)
- [PartSLIP: Low-Shot Part Segmentation for 3D Point Clouds via Pretrained Image-Language Models](https://mlsanity.com/paper/2212.01558/) (2022)
- [SAMPart3D: Segment Any Part in 3D Objects](https://mlsanity.com/paper/2411.07184/) (2024)
- [PARTFIELD: Learning 3D Feature Fields for Part Segmentation and Beyond](https://mlsanity.com/paper/2504.11451/) (2025)
- [SATR: Zero-Shot Semantic Segmentation of 3D Shapes](https://mlsanity.com/paper/2304.04909/) (2023)
- [X-Part: high fidelity and structure coherent shape decomposition](https://mlsanity.com/paper/2509.08643/) (2025)
- [PartSLIP++: Enhancing Low-Shot 3D Part Segmentation via Multi-View Instance Segmentation and Maximum Likelihood Estimation](https://mlsanity.com/paper/2312.03015/) (2023)
- [Segment Any Mesh](https://mlsanity.com/paper/2408.13679/) (2024)
- [Find Any Part in 3D](https://mlsanity.com/paper/2411.13550/) (2024)
- [PartDistill: 3D Shape Part Segmentation by Vision-Language Model Distillation](https://mlsanity.com/paper/2312.04016/) (2023)
- [PartSAM: A Scalable Promptable Part Segmentation Model Trained on Native 3D Data](https://mlsanity.com/paper/2509.21965/) (2025)
- [GeoSAM2: Unleashing the Power of SAM2 for 3D Part Segmentation](https://mlsanity.com/paper/2508.14036/) (2025)

## Abstract (arXiv)

Open-world promptable 3D semantic segmentation remains brittle as semantics are inferred in the input sensor coordinates. Yet, humans, in contrast, interpret parts via functional roles in a canonical space -- wings extend laterally, handles protrude to the side, and legs support from below. Psychophysical evidence shows that we mentally rotate objects into canonical frames to reveal these roles. To fill this gap, we propose methodName, which attains canonical space perception by inducing a latent canonical reference frame learned directly from data. By construction, we create a unified canonical dataset through LLM-guided intra- and cross-category alignment, exposing canonical spatial regularities across 200 categories. By induction, we realize canonicality inside the model through a dual-branch architecture with canonical map anchoring and canonical box calibration, collapsing pose variation and symmetry into a stable canonical embedding. This shift from input pose space to canonical embedding yields far more stable and transferable part semantics. Experimental results show that methodName establishes new state of the art in open-world promptable 3D segmentation.