# Topology Sculptor, Shape Refiner: Discrete Diffusion Model for High-Fidelity 3D Meshes Generation

arXiv: https://arxiv.org/abs/2510.21264 · HTML: https://mlsanity.com/paper/2510.21264/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in 3D, point cloud and scene generation, one of 14 directions in Diffusion models.
- Builds on 20 of its 24 references that are in the map.
- First posted on arXiv 24 Oct 2025.

## Placed in

- Diffusion models / [3D, point cloud and scene generation](https://mlsanity.com/diffusion/3d-point-cloud-and-scene-generation/)

## Builds on (most cited first, 20 of 20)

- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [ShapeNet: An Information-Rich 3D Model Repository](https://mlsanity.com/paper/1512.03012/) (2015)
- [Structured 3D Latents for Scalable and Versatile 3D Generation](https://mlsanity.com/paper/2412.01506/) (2024)
- [Wonder3D: Single Image to 3D using Cross-Domain Diffusion](https://mlsanity.com/paper/2310.15008/) (2023)
- [MeshGPT: Generating Triangle Meshes with Decoder-Only Transformers](https://mlsanity.com/paper/2311.15475/) (2023)
- [Thingi10K: A Dataset of 10,000 3D-Printing Models](https://mlsanity.com/paper/1605.04797/) (2016)
- [Seed Diffusion: A Large-Scale Diffusion Language Model with High-Speed Inference](https://mlsanity.com/paper/2508.02193/) (2025)
- [Hunyuan3D 2.1: From Images to High-Fidelity 3D Assets with Production-Ready PBR Material](https://mlsanity.com/paper/2506.15442/) (2025)
- [Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details](https://mlsanity.com/paper/2506.16504/) (2025)
- [MeshAnything V2: Artist-Created Mesh Generation With Adjacent Mesh Tokenization](https://mlsanity.com/paper/2408.02555/) (2024)
- [MeshXL: Neural Coordinate Field for Generative 3D Foundation Models](https://mlsanity.com/paper/2405.20853/) (2024)
- [EdgeRunner: Auto-regressive Auto-encoder for Artistic Mesh Generation](https://mlsanity.com/paper/2409.18114/) (2024)
- [Scaling Mesh Generation via Compressive Tokenization](https://mlsanity.com/paper/2411.07025/) (2024)
- [PolyDiff: Generating 3D Polygonal Meshes with Diffusion Models](https://mlsanity.com/paper/2312.11417/) (2023)
- [DeepMesh: Auto-Regressive Artist-mesh Creation with Reinforcement Learning](https://mlsanity.com/paper/2503.15265/) (2025)
- [Flow Matching with General Discrete Paths: A Kinetic-Optimal Perspective](https://mlsanity.com/paper/2412.03487/) (2024)
- [FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities](https://mlsanity.com/paper/2505.20147/) (2025)
- [TreeMeshGPT: Artistic Mesh Generation with Autoregressive Tree Sequencing](https://mlsanity.com/paper/2503.11629/) (2025)
- [MeshCraft: Exploring Efficient and Controllable Mesh Generation with Flow-based DiTs](https://mlsanity.com/paper/2503.23022/) (2025)
- [MeshMosaic: Scaling Artist Mesh Generation via Local-to-Global Assembly](https://mlsanity.com/paper/2509.19995/) (2025)

## Built on by (most cited first, 5 of 5)

- [PartDiffuser: Part-wise 3D Mesh Generation via Discrete Diffusion](https://mlsanity.com/paper/2511.18801/) (2025)
- [DVD: Discrete Voxel Diffusion for 3D Generation and Editing](https://mlsanity.com/paper/2605.07971/) (2026)
- [Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion](https://mlsanity.com/paper/2608.19567/) (2026)
- [QuadLink: Autoregressive Quad-Dominant Mesh Generation via Point-Relation Learning](https://mlsanity.com/paper/2605.16813/) (2026)
- [MeshOctave: Vertex Split-and-Rewire Cascades for Native Mesh Generation](https://mlsanity.com/paper/2609.38985/) (2026)

## Abstract (arXiv)

In this paper, we introduce Topology Sculptor, Shape Refiner (TSSR), a novel method for generating high-quality, artist-style 3D meshes based on Discrete Diffusion Models (DDMs). Our primary motivation for TSSR is to achieve highly accurate token prediction while enabling parallel generation, a significant advantage over sequential autoregressive methods. By allowing TSSR to "see" all mesh tokens concurrently, we unlock a new level of efficiency and control. We leverage this parallel generation capability through three key innovations: 1) Decoupled Training and Hybrid Inference, which distinctly separates the DDM-based generation into a topology sculpting stage and a subsequent shape refinement stage. This strategic decoupling enables TSSR to effectively capture both intricate local topology and overarching global shape. 2) An Improved Hourglass Architecture, featuring bidirectional attention enriched by face-vertex-sequence level Rotational Positional Embeddings (RoPE), thereby capturing richer contextual information across the mesh structure. 3) A novel Connection Loss, which acts as a topological constraint to further enhance the realism and fidelity of the generated meshes. Extensive experiments on complex datasets demonstrate that TSSR generates high-quality 3D artist-style meshes, capable of achieving up to 10,000 faces at a remarkable spatial resolution of 10243. The code will be released at: https://github.com/psky1111/Tencent-TSSR.