[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86001-en":3,"doc-seo-86001-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},86001,1374391975076,"Riley","https://ap-avatar.wpscdn.com/avatar/14000253ca4ec9f6853?x-image-process=image/resize,m_fixed,w_180,h_180&k=1783305029341752051",8,"Research & Report","LATO.2 Factorized 3D Mesh Generation with Vertex and Topology Flow","Flow matching over carefully designed latent representations has emerged as a powerful paradigm for topology-aware mesh generation, but prior methods entangle continuous vertex geometry with discrete combinatorial connectivity in a single latent space, leading to drifting vertices and broken surfaces. LATO.2 introduces a factorized flow matching framework that generates vertices via V-Flow and predicts connectivity via T-Flow conditioned on realized vertices, both anchored to a shared coarse voxel scaffold. Dedicated VAEs recover sub-voxel vertices and embed connectivity in a continuous latent space, enabling part-wise generation and topology-adaptive editing while improving geometric fidelity and connectivity quality.","arXiv :2607 . 10623v1 [ cs .GR] 12 Jul 2026  \nLATO. 2: FACTORIZED 3D MESH GENERATION WITH VERTEX AND TOPOLOGY FLOW  \nHang Long 1 ,2 ,∗ Tianhao Zhao 1 ,2 ,∗ Junkai Lin 1 ,2 Youjia Zhang 1 ,2 Huipeng Guo 1 Rendong Liang2 Jiale Xu2 Jozef Hladk´y3 Matthias Nießner4 Wei Yang 1 ,†  \n1Huazhong University of Science and Technology 2Meshy AI  \n3Independent Researcher 4Technical University of Munich  \n[https://github.com/LoHhhha/LATO.2](https://github.com/LoHhhha/LATO.2)  \nFigure 1: We present LATO.2, which factorizes mesh generation into a vertex flow (V-Flow) generating vertex positions under a controllable vertex count, and a topology flow (T-Flow) predicting connectivity from realized vertices. It supports high-quality generation (bottom left), part-wise generation at scalable resolution (bottom middle), and topology-adaptive editing (bottom right) .  \nABSTRACT  \nFlow matching over carefully designed latent representations has recently emerged as a powerful paradigm for topology-aware mesh generation. Existing approaches, however, model vertices and connectivity jointly in a joint latent space, entangling continuous vertex geometry with discrete combinatorial structure; this complicates flow learning and manifests as drifting vertices and broken surfaces. We present LATO.2 , a factorized flow matching framework that decomposes mesh generation into a vertex flow followed by a connectivity flow conditioned on the realized vertices, with both stages anchored to a shared coarse voxel scaffold. Dedicated VAEs underpin the two stages, recovering vertices at sub-voxel precision and embedding discrete connectivity into a continuous latent space. We demonstrate two advantages unique to this factorization: (i) part-wise generation, in which the scaffold is partitioned and each part synthesized at full latent capacity, yielding substantially higher-resolution meshes than a monolithic latent permits; and (ii) topology-adaptive editing, in which manipulating first-stage vertices induces the corresponding connectivity without re-optimization. Experi  \nments show that LATO.2 surpasses state-of-the-art topology-aware mesh genera tors in geometric fidelity and connectivity quality.  \n1 ∗ Equal contribution. 1,2 This work is done while interning with Meshy AI.  \n2† [Corresponding author: weiyangcs@hust.edu.cn](Corresponding author: weiyangcs@hust.edu.cn).  \n1 INTRODUCTION  \nArtist-created meshes exhibit compact, well-structured topology, characterized by adaptive vertex placement and clean, coherent face connectivity. Such structure is essential in production pipelines, where it supports reliable rigging and deformation for animation, high-quality shading, and efficient storage and rendering. Yet while topology-agnostic 3D generation has advanced rapidly, whether through VecSet representations (Zhang et al., 2023 ; 2024 ; Zhao et al., 2025b ; Lai et al., 2026) or structured latents (Xiang et al., 2025 ; Wu et al., 2026 ; Li et al., 2026b), these pipelines produce geometry as neural fields and extract surfaces via iso-surfacing, yielding dense, irregular tessellations devoid of artist-like structure. Directly generating meshes with such structure remains difficult because meshes are inherently hybrid representations: vertex coordinates are coupled with discrete, combinatorial connectivity of variable cardinality, not naturally captured by conventional generative models formulated over fixed-dimensional continuous spaces. To confront this discreteness, a prominent line of work tokenizes meshes into sequences and trains autoregressive models to generate faces token by token, often conditioned on point clouds (Siddiqui et al., 2024 ; Chen et al., 2024 ; 2025b ; Hao et al., 2024 ; Weng et al., 2025) . By modeling connectivity through sequential token prediction, these approaches produce structurally coherent, production-oriented meshes. However, serializing large, detailed meshes yields prohibitively long token streams, which inflates training cost, slows","cbCaicEt4JnhZAw6","https://ap.wps.com/l/cbCaicEt4JnhZAw6","pdf",41788826,3,1,18,"English","en",105,"# Abstract\n# Introduction\n## Motivation: limitations of unified latent modeling\n## Prior directions in mesh representations and generation\n## Flow-based topology-aware generation and related methods\n## Core idea of LATO.2 factorization","[{\"question\":\"What problem does LATO.2 address in topology-aware mesh generation?\",\"answer\":\"It addresses the entanglement of continuous vertex geometry with discrete connectivity when both are modeled in a single joint latent space, which complicates flow learning and can produce drifting vertices and broken surfaces.\"},{\"question\":\"How does LATO.2 factorize the mesh generation process?\",\"answer\":\"LATO.2 decomposes generation into a vertex flow (V-Flow) that produces vertex positions under a controllable vertex count and a topology flow (T-Flow) that predicts connectivity conditioned on the realized vertices.\"},{\"question\":\"What advantages does the factorized design enable?\",\"answer\":\"It enables part-wise generation at scalable resolution with higher mesh detail, and topology-adaptive editing where modifying first-stage vertices induces the corresponding connectivity without re-optimization.\"}]",1784207688,45,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"lato2-factorized-3d-mesh-generation-with-vertex-and-topology-flow","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":20},"https://docshare.wps.com/document/research-report/",{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/lato2-factorized-3d-mesh-generation-with-vertex-and-topology-flow/86001/",4,{"url":51,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does LATO.2 address in topology-aware mesh generation?","Question",{"text":75,"@type":76},"It addresses the entanglement of continuous vertex geometry with discrete connectivity when both are modeled in a single joint latent space, which complicates flow learning and can produce drifting vertices and broken surfaces.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does LATO.2 factorize the mesh generation process?",{"text":80,"@type":76},"LATO.2 decomposes generation into a vertex flow (V-Flow) that produces vertex positions under a controllable vertex count and a topology flow (T-Flow) that predicts connectivity conditioned on the realized vertices.",{"name":82,"@type":73,"acceptedAnswer":83},"What advantages does the factorized design enable?",{"text":84,"@type":76},"It enables part-wise generation at scalable resolution with higher mesh detail, and topology-adaptive editing where modifying first-stage vertices induces the corresponding connectivity without re-optimization.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]