{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:N5YKGWFZCTE7PYUWVBCYKIYP2E","short_pith_number":"pith:N5YKGWFZ","schema_version":"1.0","canonical_sha256":"6f70a358b914c9f7e296a84585230fd11f20a91bb4452fe1e13b94e558dc40d8","source":{"kind":"arxiv","id":"2412.01786","version":2},"attestation_state":"computed","paper":{"title":"Gradient-Free Generation for Hard-Constrained Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abdul Fatir Ansari, Andrew Stuart, Boran Han, Chaoran Cheng, Danielle C. Maddix, Michael W. Mahoney, Yuyang Wang","submitted_at":"2024-12-02T18:36:26Z","abstract_excerpt":"Generative models that satisfy hard constraints are critical in many scientific and engineering applications, where physical laws or system requirements must be strictly respected. Many existing constrained generative models, especially those developed for computer vision, rely heavily on gradient information, which is often sparse or computationally expensive in some fields, e.g., partial differential equations (PDEs). In this work, we introduce a novel framework for adapting pre-trained, unconstrained flow-matching models to satisfy constraints exactly in a zero-shot manner without requiring"},"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":"2412.01786","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-02T18:36:26Z","cross_cats_sorted":[],"title_canon_sha256":"d3c591a6d904dec3153e8289024d6017074e8018837cc7ce63a08e0301233661","abstract_canon_sha256":"c03dd3da05972709aa9f5d72b5e2abc83d0cb14a4aad62a4cfc07ff12f091de0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:23:27.361655Z","signature_b64":"jnKpSi0ibHkpsMOJnzQr20KNBJDogDfWYfLJ/enWbPrhJvH6ZGT2bswUNTd9TBibkiL5MHlY8Q6cS/chgS2RAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f70a358b914c9f7e296a84585230fd11f20a91bb4452fe1e13b94e558dc40d8","last_reissued_at":"2026-07-05T10:23:27.361148Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:23:27.361148Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Gradient-Free Generation for Hard-Constrained Systems","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Abdul Fatir Ansari, Andrew Stuart, Boran Han, Chaoran Cheng, Danielle C. Maddix, Michael W. Mahoney, Yuyang Wang","submitted_at":"2024-12-02T18:36:26Z","abstract_excerpt":"Generative models that satisfy hard constraints are critical in many scientific and engineering applications, where physical laws or system requirements must be strictly respected. Many existing constrained generative models, especially those developed for computer vision, rely heavily on gradient information, which is often sparse or computationally expensive in some fields, e.g., partial differential equations (PDEs). In this work, we introduce a novel framework for adapting pre-trained, unconstrained flow-matching models to satisfy constraints exactly in a zero-shot manner without requiring"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.01786","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/2412.01786/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":"2412.01786","created_at":"2026-07-05T10:23:27.361210+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.01786v2","created_at":"2026-07-05T10:23:27.361210+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.01786","created_at":"2026-07-05T10:23:27.361210+00:00"},{"alias_kind":"pith_short_12","alias_value":"N5YKGWFZCTE7","created_at":"2026-07-05T10:23:27.361210+00:00"},{"alias_kind":"pith_short_16","alias_value":"N5YKGWFZCTE7PYUW","created_at":"2026-07-05T10:23:27.361210+00:00"},{"alias_kind":"pith_short_8","alias_value":"N5YKGWFZ","created_at":"2026-07-05T10:23:27.361210+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12754","citing_title":"Constraint-Aware Flow Matching: Decision Aligned End-to-End Training for Constrained Sampling","ref_index":20,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E","json":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E.json","graph_json":"https://pith.science/api/pith-number/N5YKGWFZCTE7PYUWVBCYKIYP2E/graph.json","events_json":"https://pith.science/api/pith-number/N5YKGWFZCTE7PYUWVBCYKIYP2E/events.json","paper":"https://pith.science/paper/N5YKGWFZ"},"agent_actions":{"view_html":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E","download_json":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E.json","view_paper":"https://pith.science/paper/N5YKGWFZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.01786&json=true","fetch_graph":"https://pith.science/api/pith-number/N5YKGWFZCTE7PYUWVBCYKIYP2E/graph.json","fetch_events":"https://pith.science/api/pith-number/N5YKGWFZCTE7PYUWVBCYKIYP2E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E/action/storage_attestation","attest_author":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E/action/author_attestation","sign_citation":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E/action/citation_signature","submit_replication":"https://pith.science/pith/N5YKGWFZCTE7PYUWVBCYKIYP2E/action/replication_record"}},"created_at":"2026-07-05T10:23:27.361210+00:00","updated_at":"2026-07-05T10:23:27.361210+00:00"}