High-fidelity 3D meshes can be tokenized into one-dimension (1D) sequences and directly modeled using autoregressive approaches for faces and vertices. However, existing methods suffer from insufficient resource utilization, resulting in slow inference and the ability to handle only small-scale sequences, which severely constrains the expressible structural details. We introduce the Latent Autoregressive Network (LANE), which incorporates compact autoregressive dependencies in the generation process, achieving a $6\times$ improvement in maximum generatable sequence length compared to existing methods. To further accelerate inference, we propose the Adaptive Computation Graph Reconfiguration (AdaGraph) strategy, which effectively overcomes the efficiency bottleneck of traditional serial inference through spatiotemporal decoupling in the generation process. Experimental validation demonstrates that LANE achieves superior performance across generation speed, structural detail, and geometric consistency, providing an effective solution for high-quality 3D mesh generation.

翻译：高保真三维网格可被分词化为一维序列，并可直接通过自回归方法对面和顶点进行建模。然而，现有方法存在资源利用不足的问题，导致推理速度缓慢，且仅能处理小规模序列，这严重限制了可表达的结构细节。我们引入了潜在自回归网络（LANE），该网络在生成过程中融入了紧凑的自回归依赖关系，与现有方法相比，可实现最大可生成序列长度$6\times$的提升。为了进一步加速推理，我们提出了自适应计算图重构（AdaGraph）策略，该策略通过生成过程中的时空解耦，有效克服了传统串行推理的效率瓶颈。实验验证表明，LANE在生成速度、结构细节和几何一致性方面均实现了优越性能，为高质量三维网格生成提供了有效解决方案。