Related papers: FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation

FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation

URL: http://arxiv.org/abs/2511.15618v1
Date: Wed, 19 Nov 2025 17:03:49 GMT
Title: FlashMesh: Faster and Better Autoregressive Mesh Synthesis via Structured Speculation
Authors: Tingrui Shen, Yiheng Zhang, Chen Tang, Chuan Ping, Zixing Zhao, Le Wan, Yuwang Wang, Ronggang Wang, Shengfeng He,
Abstract summary: FlashMesh is a fast and high-fidelity mesh generation framework.<n>We show that FlashMesh achieves up to a 2 x speedup over standard autoregressive models.
Score: 65.3277633028397
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Abstract: Autoregressive models can generate high-quality 3D meshes by sequentially producing vertices and faces, but their token-by-token decoding results in slow inference, limiting practical use in interactive and large-scale applications. We present FlashMesh, a fast and high-fidelity mesh generation framework that rethinks autoregressive decoding through a predict-correct-verify paradigm. The key insight is that mesh tokens exhibit strong structural and geometric correlations that enable confident multi-token speculation. FlashMesh leverages this by introducing a speculative decoding scheme tailored to the commonly used hourglass transformer architecture, enabling parallel prediction across face, point, and coordinate levels. Extensive experiments show that FlashMesh achieves up to a 2 x speedup over standard autoregressive models while also improving generation fidelity. Our results demonstrate that structural priors in mesh data can be systematically harnessed to accelerate and enhance autoregressive generation.

Related papers

FACE: A Face-based Autoregressive Representation for High-Fidelity and Efficient Mesh Generation [50.71369329585773]
We introduce FACE, a novel Autoregressive Autoencoder framework that generates meshes at the face level.<n>Our one-face-one-token strategy treats each triangle face, the fundamental building block of a mesh, as a single, unified token.<n> FACE achieves state-of-the-art reconstruction quality on standard benchmarks.
arXiv Detail & Related papers (2026-03-02T06:47:15Z)
HiFi-Mesh: High-Fidelity Efficient 3D Mesh Generation via Compact Autoregressive Dependence [36.403921772528236]
We introduce the Latent Autoregressive Network (LANE), which incorporates compact autoregressive dependencies in the generation process.<n>LANE achieves a $6times$ improvement in maximum sequence length compared to existing methods.
arXiv Detail & Related papers (2026-01-29T06:22:26Z)
Auto-Regressive Masked Diffusion Models [9.239507801466322]
Masked diffusion models (MDMs) have emerged as a promising approach for language modeling.<n>They face a performance gap compared to autoregressive models (ARMs) and require more training iterations.<n>We present the Auto-Regressive Masked Diffusion model, which unifies the training efficiency of autoregressive models with the parallel generation capabilities of diffusion-based models.
arXiv Detail & Related papers (2026-01-23T18:42:30Z)
FastMesh: Efficient Artistic Mesh Generation via Component Decoupling [27.21354509059262]
Mesh generation approaches typically tokenize triangle meshes into sequences of tokens and train autoregressive models to generate these tokens sequentially.<n>This redundancy leads to excessively long token sequences and inefficient generation processes.<n>We propose an efficient framework that generates artistic meshes by treating vertices and faces separately.
arXiv Detail & Related papers (2025-08-26T16:51:02Z)
AutoHFormer: Efficient Hierarchical Autoregressive Transformer for Time Series Prediction [36.239648954658534]
Time series forecasting requires architectures that simultaneously achieve three competing objectives.<n>We introduce AutoHFormer, a hierarchical autoregressive transformer that addresses these challenges.<n> Comprehensive experiments demonstrate that AutoHFormer 10.76X faster training and 6.06X memory reduction compared to PatchTST on P08.
arXiv Detail & Related papers (2025-06-19T03:47:04Z)
Fast Autoregressive Models for Continuous Latent Generation [49.079819389916764]
Autoregressive models have demonstrated remarkable success in sequential data generation, particularly in NLP.<n>Recent work, the masked autoregressive model (MAR) bypasses quantization by modeling per-token distributions in continuous spaces using a diffusion head.<n>We propose Fast AutoRegressive model (FAR), a novel framework that replaces MAR's diffusion head with a lightweight shortcut head.
arXiv Detail & Related papers (2025-04-24T13:57:08Z)
MeshCraft: Exploring Efficient and Controllable Mesh Generation with Flow-based DiTs [79.45006864728893]
MeshCraft is a framework for efficient and controllable mesh generation.<n>It uses continuous spatial diffusion to generate discrete triangle faces.<n>It can generate an 800-face mesh in just 3.2 seconds.
arXiv Detail & Related papers (2025-03-29T09:21:50Z)
Parallelized Autoregressive Visual Generation [65.9579525736345]
We propose a simple yet effective approach for parallelized autoregressive visual generation.<n>Our method achieves a 3.6x speedup with comparable quality and up to 9.5x speedup with minimal quality degradation across both image and video generation tasks.
arXiv Detail & Related papers (2024-12-19T17:59:54Z)
Promises and Pitfalls of Generative Masked Language Modeling: Theoretical Framework and Practical Guidelines [74.42485647685272]
We focus on Generative Masked Language Models (GMLMs) We train a model to fit conditional probabilities of the data distribution via masking, which are subsequently used as inputs to a Markov Chain to draw samples from the model. We adapt the T5 model for iteratively-refined parallel decoding, achieving 2-3x speedup in machine translation with minimal sacrifice in quality.
arXiv Detail & Related papers (2024-07-22T18:00:00Z)

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