Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion

Hugging Face Daily Papers Papers

Summary

Block3D accelerates text-to-3D generation by using block-wise diffusion with confidence-guided correction to reduce inference time while preserving geometric fidelity, achieving a 5.15x speedup.

While text-to-3D generation has advanced rapidly, achieving high geometric fidelity at low inference cost remains challenging. Existing text-to-3D methods either decode discrete shape tokens autoregressively or iteratively refine global 3D representations with diffusion or flow models. However, autoregressive decoding is sequential and cannot revise errors, whereas diffusion and flow-matching models repeatedly process the full representation, making high-quality generation increasingly expensive. In this paper, we propose Block3D, a block-wise diffusion framework that partitions the discrete shape-token sequence into contiguous blocks, generates the blocks autoregressively, and jointly denoises all tokens within the current block. To alleviate error accumulation, we introduce confidence-guided intra-block correction, which revises low-confidence tokens before each block is finalized. On a held-out set from TRELLIS-500K, Block3D reduces mean end-to-end generation time from 25.71 seconds to 4.99 seconds, achieving a 5.15times speedup over the fine-tuned autoregressive baseline without sacrificing geometric fidelity.
Original Article
View Cached Full Text

Cached at: 08/25/26, 08:35 AM

Paper page - Block3D: Efficient Text-to-3D Generation via Block-Wise Diffusion

Source: https://huggingface.co/papers/2608.19567

Abstract

Block3D accelerates text-to-3D generation by using block-wise diffusion with confidence-guided correction to reduce inference time while preserving geometric fidelity.

Whiletext-to-3D generationhas advanced rapidly, achieving high geometric fidelity at low inference cost remains challenging. Existing text-to-3D methods either decodediscrete shape tokensautoregressively or iteratively refine global 3D representations with diffusion or flow models. However,autoregressive decodingis sequential and cannot revise errors, whereas diffusion andflow-matchingmodels repeatedly process the full representation, making high-quality generation increasingly expensive. In this paper, we propose Block3D, ablock-wise diffusionframework that partitions the discrete shape-token sequence into contiguous blocks, generates the blocks autoregressively, and jointly denoises all tokens within the current block. To alleviate error accumulation, we introduceconfidence-guided intra-block correction, which revises low-confidence tokens before each block is finalized. On a held-out set from TRELLIS-500K, Block3D reduces mean end-to-end generation time from 25.71 seconds to 4.99 seconds, achieving a 5.15times speedup over the fine-tuned autoregressive baseline without sacrificing geometric fidelity.

View arXiv pageView PDFProject pageGitHub3Add to collection

Get this paper in your agent:

hf papers read 2608\.19567

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2608.19567 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2608.19567 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2608.19567 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

DiffusionGemma: 4x Faster Text Generation

Hacker News Top

Google introduces DiffusionGemma, an experimental 26B MoE open model that achieves up to 4x faster text generation on GPUs using text diffusion, targeting speed-critical interactive local workflows.

Flash-BoN: Instant Drafts for Inference-Time Scaling in Diffusion Models

Hugging Face Daily Papers

Flash-BoN improves text-to-image generation efficiency by generating cheap draft candidates via timestep truncation, layer skipping, and activation proxies, then using multi-stage verification to select the best draft for full refinement, outperforming baselines under fixed wall-clock budgets.