{"ID":2831682,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.07514","arxiv_id":"2512.07514","title":"MeshRipple: Structured Autoregressive Generation of Artist-Meshes","abstract":"Meshes serve as a primary representation for 3D assets. Autoregressive mesh generators serialize faces into sequences and train on truncated segments with sliding-window inference to cope with memory limits. However, this mismatch breaks long-range geometric dependencies, producing holes and fragmented components. To address this critical limitation, we introduce MeshRipple, which expands a mesh outward from an active generation frontier, akin to a ripple on a surface. MeshRipple rests on three key innovations: a frontier-aware BFS tokenization that aligns the generation order with surface topology; an expansive prediction strategy that maintains coherent, connected surface growth; and a sparse-attention global memory that provides an effectively unbounded receptive field to resolve long-range topological dependencies. This integrated design enables MeshRipple to generate meshes with high surface fidelity and topological completeness, outperforming strong recent baselines.","short_abstract":"Meshes serve as a primary representation for 3D assets. Autoregressive mesh generators serialize faces into sequences and train on truncated segments with sliding-window inference to cope with memory limits. However, this mismatch breaks long-range geometric dependencies, producing holes and fragmented components. To a...","url_abs":"https://arxiv.org/abs/2512.07514","url_pdf":"https://arxiv.org/pdf/2512.07514v2","authors":"[\"Junkai Lin\",\"Hang Long\",\"Huipeng Guo\",\"Jielei Zhang\",\"JiaYi Yang\",\"Tianle Guo\",\"Yang Yang\",\"Jianwen Li\",\"Wenxiao Zhang\",\"Matthias Nießner\",\"Wei Yang\"]","published":"2025-12-08T12:50:23Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[]","has_code":false}