{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:FC3KD7LA3KSO2XJC3N6KL6MPJV","short_pith_number":"pith:FC3KD7LA","schema_version":"1.0","canonical_sha256":"28b6a1fd60daa4ed5d22db7ca5f98f4d4dfaa3d73a3054f4ac93d229e7ce0de7","source":{"kind":"arxiv","id":"2503.16302","version":2},"attestation_state":"computed","paper":{"title":"Unleashing Vecset Diffusion Model for Fast Shape Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.IV"],"primary_cat":"cs.CV","authors_text":"Chunchao Guo, Fuyun Wang, Haolin Liu, Huiwen Shi, Jie Jiang, Jingwei Huang, Qingxiang Lin, Xianghui Yang, Xiangyu Yue, Yuhong Liu, Yunfei Zhao, Zeqiang Lai, Zibo Zhao","submitted_at":"2025-03-20T16:23:44Z","abstract_excerpt":"3D shape generation has greatly flourished through the development of so-called \"native\" 3D diffusion, particularly through the Vecset Diffusion Model (VDM). While recent advancements have shown promising results in generating high-resolution 3D shapes, VDM still struggles with high-speed generation. Challenges exist because of difficulties not only in accelerating diffusion sampling but also VAE decoding in VDM, areas under-explored in previous works. To address these challenges, we present FlashVDM, a systematic framework for accelerating both VAE and DiT in VDM. For DiT, FlashVDM enables fl"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.16302","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-03-20T16:23:44Z","cross_cats_sorted":["cs.AI","eess.IV"],"title_canon_sha256":"621b3cbaa5278871cd35d3da2d99cf7d2cdebc3eb550056ab4beed7287107b8e","abstract_canon_sha256":"91f92ec70ca216873a03a66edc91512058d88588b21a0b298fcac07b121a411d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:39:27.929093Z","signature_b64":"akzEb4EsIoJcH2QlytZoG/sa8Hk5Lkam7QrHNsiHKkgPi5mHxdZ/YItTMaijqUUXpf2+dGePSoKDarIi+MCIBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"28b6a1fd60daa4ed5d22db7ca5f98f4d4dfaa3d73a3054f4ac93d229e7ce0de7","last_reissued_at":"2026-07-05T10:39:27.928551Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:39:27.928551Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unleashing Vecset Diffusion Model for Fast Shape Generation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","eess.IV"],"primary_cat":"cs.CV","authors_text":"Chunchao Guo, Fuyun Wang, Haolin Liu, Huiwen Shi, Jie Jiang, Jingwei Huang, Qingxiang Lin, Xianghui Yang, Xiangyu Yue, Yuhong Liu, Yunfei Zhao, Zeqiang Lai, Zibo Zhao","submitted_at":"2025-03-20T16:23:44Z","abstract_excerpt":"3D shape generation has greatly flourished through the development of so-called \"native\" 3D diffusion, particularly through the Vecset Diffusion Model (VDM). While recent advancements have shown promising results in generating high-resolution 3D shapes, VDM still struggles with high-speed generation. Challenges exist because of difficulties not only in accelerating diffusion sampling but also VAE decoding in VDM, areas under-explored in previous works. To address these challenges, we present FlashVDM, a systematic framework for accelerating both VAE and DiT in VDM. For DiT, FlashVDM enables fl"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.16302","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.16302/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.16302","created_at":"2026-07-05T10:39:27.928615+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.16302v2","created_at":"2026-07-05T10:39:27.928615+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.16302","created_at":"2026-07-05T10:39:27.928615+00:00"},{"alias_kind":"pith_short_12","alias_value":"FC3KD7LA3KSO","created_at":"2026-07-05T10:39:27.928615+00:00"},{"alias_kind":"pith_short_16","alias_value":"FC3KD7LA3KSO2XJC","created_at":"2026-07-05T10:39:27.928615+00:00"},{"alias_kind":"pith_short_8","alias_value":"FC3KD7LA","created_at":"2026-07-05T10:39:27.928615+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.16504","citing_title":"Hunyuan3D 2.5: Towards High-Fidelity 3D Assets Generation with Ultimate Details","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"2601.11194","citing_title":"ATATA: One Algorithm to Align Them All","ref_index":30,"is_internal_anchor":false},{"citing_arxiv_id":"2605.13862","citing_title":"Seed3D 2.0: Advancing High-Fidelity Simulation-Ready 3D Content Generation","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17472","citing_title":"UniMesh: Unifying 3D Mesh Understanding and Generation","ref_index":22,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV","json":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV.json","graph_json":"https://pith.science/api/pith-number/FC3KD7LA3KSO2XJC3N6KL6MPJV/graph.json","events_json":"https://pith.science/api/pith-number/FC3KD7LA3KSO2XJC3N6KL6MPJV/events.json","paper":"https://pith.science/paper/FC3KD7LA"},"agent_actions":{"view_html":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV","download_json":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV.json","view_paper":"https://pith.science/paper/FC3KD7LA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.16302&json=true","fetch_graph":"https://pith.science/api/pith-number/FC3KD7LA3KSO2XJC3N6KL6MPJV/graph.json","fetch_events":"https://pith.science/api/pith-number/FC3KD7LA3KSO2XJC3N6KL6MPJV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV/action/storage_attestation","attest_author":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV/action/author_attestation","sign_citation":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV/action/citation_signature","submit_replication":"https://pith.science/pith/FC3KD7LA3KSO2XJC3N6KL6MPJV/action/replication_record"}},"created_at":"2026-07-05T10:39:27.928615+00:00","updated_at":"2026-07-05T10:39:27.928615+00:00"}