Search papers, labs, and topics across Lattice.
This paper introduces Block3D, a novel block-wise diffusion framework for text-to-3D generation that enhances efficiency by partitioning discrete shape-token sequences into contiguous blocks. By generating these blocks autoregressively and applying confidence-guided intra-block correction, Block3D significantly reduces error accumulation and improves generation speed. The method achieves a remarkable $5.15\times$ speedup in end-to-end generation time, decreasing it from 25.71 seconds to 4.99 seconds while maintaining high geometric fidelity on the TRELLIS-500K dataset.
Achieving a $5.15\times$ speedup in text-to-3D generation without sacrificing quality could revolutionize real-time 3D content creation.
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 $5.15\times$ speedup over the fine-tuned autoregressive baseline without sacrificing geometric fidelity.
