Block3D：透過分塊擴散實現高效文字轉3D生成

摘要

雖然文字轉 3D 生成技術已快速進展，但在低推理成本下達成高幾何保真度仍具挑戰性。現有的文字轉 3D 方法，要麼以自迴歸方式逐一解碼離散形狀 token，要麼藉由擴散或流模型反覆精化整體 3D 表示。然而，自迴歸解碼具序列性且無法修正錯誤，而擴散與流匹配模型則會反覆處理完整的表示，使得高品質生成日益昂貴。在本文中，我們提出 Block3D，一種區塊式擴散框架，其將離散形狀 token 序列分割為連續區塊，以自迴歸方式生成各區塊，並對當前區塊內的所有 token 進行聯合去噪。為減輕錯誤累積，我們引入信心引導的區塊內修正機制，在每個區塊最終確定之前修訂低信心度的 token。在 TRELLIS-500K 的保留集上，Block3D 將平均端到端生成時間從 25.71 秒降至 4.99 秒，相較於微調後的自迴歸基線取得 5.15 倍加速，且不犧牲幾何保真度。

English

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.