MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling

• URL: http://arxiv.org/abs/2608.14783v1
• Date: Fri, 14 Aug 2026 18:00:02 GMT
• Title: MegaParts: Scaling Part-Aware 3D Object Generation to 300 Parts via Token-Efficient Autoregressive Modeling
• Abstract summary: MegaParts is a scalable autoregressive 3D generation framework.<n>It scales to objects with up to 300 parts and sequence lengths up to 256k tokens.<n>Our method achieves higher mesh quality than baseline autoregressive and diffusion models.
• Score: 48.6816265114274
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Part-aware 3D object generation is essential for graphics applications such as controllable modeling, editing, and articulation, where objects are represented as coherent assemblies of semantic parts. However, existing part-aware generation methods, do not scale well to highly complex objects. As the number of parts increases, generating detailed geometry becomes prohibitively expensive in token length and memory. We introduce MegaParts, a scalable autoregressive 3D generation framework to address this challenge by combining structured sequence modeling with a token-efficient vector-quantized shape tokenizer. Our tokenizer learns discrete latent representations for part-level geometry by minimizing token usage subject to high-fidelity reconstruction, enabling adaptive-length tokenization based on geometric complexity. On top of this compact representation, we train a large language model to generate object bounding boxes, part bounding boxes, and part shape tokens within a unified structured sequence. Combined with efficient long-context training strategy, our token-efficient formulation scales to objects with up to 300 parts and sequence lengths up to 256k tokens. This substantially extends the scale of part-aware 3D generation while preserving compositional structure and enabling fine-grained part-level control. Our method achieves higher mesh quality than baseline autoregressive and diffusion models, showing that compressed discrete part tokens improve not only scalability but also the achievable fidelity of generated geometry. These results suggest that LLM native token-efficient autoregressive modeling is a compelling alternative to diffusion for large-scale part-aware 3D generation. The project page is available at https://expmaster.github.io/megaparts_webpage.

Related papers

• LoST: Level of Semantics Tokenization for 3D Shapes [50.847769883816085]
  State-of-the-art methods rely on geometric level-of-detail (LoD) hierarchies, originally designed for rendering and compression.<n>We propose Level-of-Semantics Tokenization (LoST), which orders salience, such that early prefixes decode complete, plausible shapes.<n>LoST achieves SOTA reconstruction, surpassing previous LoD-based 3D shape tokenizers by large margins on both geometric and semantic reconstruction metrics.
  arXiv  Detail & Related papers  (2026-03-18T17:56:06Z)
• ArtLLM: Generating Articulated Assets via 3D LLM [19.814132638278547]
  ArtLLM is a novel framework for generating high-quality articulated assets directly from complete 3D meshes.<n>At its core is a 3D multimodal large language model trained on a large-scale articulation dataset.<n> Experiments show that ArtLLM significantly outperforms state-of-the-art methods in both part layout accuracy and joint prediction.
  arXiv  Detail & Related papers  (2026-03-01T15:07:46Z)
• Part-X-MLLM: Part-aware 3D Multimodal Large Language Model [35.75184591224847]
  Part-X-MLLM is a native 3D multimodal large language model.<n>It unifies diverse 3D tasks by formulating them as programs in a structured, executable grammar.
  arXiv  Detail & Related papers  (2025-11-17T17:59:52Z)
• PartSAM: A Scalable Promptable Part Segmentation Model Trained on Native 3D Data [47.60227259482637]
  We present PartSAM, the first promptable part segmentation model trained on large-scale 3D data.<n>PartSAM employs an encoder-decoder architecture in which a triplane-based dual-branch encoder produces spatially structured tokens.<n>To enable large-scale supervision, we introduce a model-in-the-loop annotation pipeline that curates over five million 3D shape-part pairs.
  arXiv  Detail & Related papers  (2025-09-26T06:52:35Z)
• X-Part: high fidelity and structure coherent shape decomposition [21.08914191612629]
  X-Part is a controllable generative model designed to decompose a holistic 3D object into semantically meaningful parts.<n>X-Part exploits the bounding box as prompts for the part generation and injects point-wise semantic features for meaningful decomposition.<n>This work establishes a new paradigm for creating production-ready, editable, and structurally sound 3D assets.
  arXiv  Detail & Related papers  (2025-09-10T14:37:02Z)
• MeshLLM: Empowering Large Language Models to Progressively Understand and Generate 3D Mesh [79.20802127426003]
  MeshLLM is a framework that leverages large language models (LLMs) to understand and generate text-serialized 3D meshes.<n>We introduce a Primitive-Mesh decomposition strategy, which divides 3D meshes into structurally meaningful subunits.<n> Experiments show that MeshLLM outperforms the state-of-the-art LLaMA-Mesh in both mesh generation quality and shape understanding.
  arXiv  Detail & Related papers  (2025-08-02T07:37:37Z)
• 3D Part Segmentation via Geometric Aggregation of 2D Visual Features [57.20161517451834]
  Supervised 3D part segmentation models are tailored for a fixed set of objects and parts, limiting their transferability to open-set, real-world scenarios.<n>Recent works have explored vision-language models (VLMs) as a promising alternative, using multi-view rendering and textual prompting to identify object parts.<n>To address these limitations, we propose COPS, a COmprehensive model for Parts that blends semantics extracted from visual concepts and 3D geometry to effectively identify object parts.
  arXiv  Detail & Related papers  (2024-12-05T15:27:58Z)
• Pushing Auto-regressive Models for 3D Shape Generation at Capacity and Scalability [118.26563926533517]
  Auto-regressive models have achieved impressive results in 2D image generation by modeling joint distributions in grid space. We extend auto-regressive models to 3D domains, and seek a stronger ability of 3D shape generation by improving auto-regressive models at capacity and scalability simultaneously.
  arXiv  Detail & Related papers  (2024-02-19T15:33:09Z)
• Learning Versatile 3D Shape Generation with Improved AR Models [91.87115744375052]
  Auto-regressive (AR) models have achieved impressive results in 2D image generation by modeling joint distributions in the grid space. We propose the Improved Auto-regressive Model (ImAM) for 3D shape generation, which applies discrete representation learning based on a latent vector instead of volumetric grids.
  arXiv  Detail & Related papers  (2023-03-26T12:03:18Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.