{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:V77NVS3ELBTE46HP25Z56RIG6E","short_pith_number":"pith:V77NVS3E","schema_version":"1.0","canonical_sha256":"affedacb6458664e78efd773df4506f1354650f093e0a32562604eb17b44f98f","source":{"kind":"arxiv","id":"2504.20179","version":1},"attestation_state":"computed","paper":{"title":"Integration Flow Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dan Zhang, Feng Luo, Jingjing Wang, Joshua Luo, Yin Yang","submitted_at":"2025-04-28T18:29:15Z","abstract_excerpt":"Ordinary differential equation (ODE) based generative models have emerged as a powerful approach for producing high-quality samples in many applications. However, the ODE-based methods either suffer the discretization error of numerical solvers of ODE, which restricts the quality of samples when only a few NFEs are used, or struggle with training instability. In this paper, we proposed Integration Flow, which directly learns the integral of ODE-based trajectory paths without solving the ODE functions. Moreover, Integration Flow explicitly incorporates the target state $\\mathbf{x}_0$ as the anc"},"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":"2504.20179","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2025-04-28T18:29:15Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"7b74128a351dead583a4847f49f6561c92656d5119ae7578df4b517ee7c7c112","abstract_canon_sha256":"46a604df357d0bfa03478f91789346c887ee6ed576eee822a9eaa4559ae54120"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:55:37.928207Z","signature_b64":"NC3vb6c7T3G7hua65GV4uF+r2VhDIEjkMsrc8c1/I1c18sVYc7t41CnOIXCN6qhrGaZGbLS3KHfowjhJz77wBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"affedacb6458664e78efd773df4506f1354650f093e0a32562604eb17b44f98f","last_reissued_at":"2026-07-05T10:55:37.927721Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:55:37.927721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Integration Flow Models","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Dan Zhang, Feng Luo, Jingjing Wang, Joshua Luo, Yin Yang","submitted_at":"2025-04-28T18:29:15Z","abstract_excerpt":"Ordinary differential equation (ODE) based generative models have emerged as a powerful approach for producing high-quality samples in many applications. However, the ODE-based methods either suffer the discretization error of numerical solvers of ODE, which restricts the quality of samples when only a few NFEs are used, or struggle with training instability. In this paper, we proposed Integration Flow, which directly learns the integral of ODE-based trajectory paths without solving the ODE functions. Moreover, Integration Flow explicitly incorporates the target state $\\mathbf{x}_0$ as the anc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2504.20179","kind":"arxiv","version":1},"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/2504.20179/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":"2504.20179","created_at":"2026-07-05T10:55:37.927779+00:00"},{"alias_kind":"arxiv_version","alias_value":"2504.20179v1","created_at":"2026-07-05T10:55:37.927779+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2504.20179","created_at":"2026-07-05T10:55:37.927779+00:00"},{"alias_kind":"pith_short_12","alias_value":"V77NVS3ELBTE","created_at":"2026-07-05T10:55:37.927779+00:00"},{"alias_kind":"pith_short_16","alias_value":"V77NVS3ELBTE46HP","created_at":"2026-07-05T10:55:37.927779+00:00"},{"alias_kind":"pith_short_8","alias_value":"V77NVS3E","created_at":"2026-07-05T10:55:37.927779+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.17426","citing_title":"Modular MeanFlow: Towards Stable and Scalable One-Step Generative Modeling","ref_index":38,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E","json":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E.json","graph_json":"https://pith.science/api/pith-number/V77NVS3ELBTE46HP25Z56RIG6E/graph.json","events_json":"https://pith.science/api/pith-number/V77NVS3ELBTE46HP25Z56RIG6E/events.json","paper":"https://pith.science/paper/V77NVS3E"},"agent_actions":{"view_html":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E","download_json":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E.json","view_paper":"https://pith.science/paper/V77NVS3E","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2504.20179&json=true","fetch_graph":"https://pith.science/api/pith-number/V77NVS3ELBTE46HP25Z56RIG6E/graph.json","fetch_events":"https://pith.science/api/pith-number/V77NVS3ELBTE46HP25Z56RIG6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E/action/storage_attestation","attest_author":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E/action/author_attestation","sign_citation":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E/action/citation_signature","submit_replication":"https://pith.science/pith/V77NVS3ELBTE46HP25Z56RIG6E/action/replication_record"}},"created_at":"2026-07-05T10:55:37.927779+00:00","updated_at":"2026-07-05T10:55:37.927779+00:00"}