{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:DXWMR2H2C4VVFJ7H22FHMFZYSL","short_pith_number":"pith:DXWMR2H2","schema_version":"1.0","canonical_sha256":"1decc8e8fa172b52a7e7d68a76173892ea6a1cf5f4cda6ca291136c8c79333c2","source":{"kind":"arxiv","id":"2506.09258","version":1},"attestation_state":"computed","paper":{"title":"CFMI: Flow Matching for Missing Data Imputation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Michael U. Gutmann, Vaidotas Simkus","submitted_at":"2025-06-10T21:40:36Z","abstract_excerpt":"We introduce conditional flow matching for imputation (CFMI), a new general-purpose method to impute missing data. The method combines continuous normalising flows, flow-matching, and shared conditional modelling to deal with intractabilities of traditional multiple imputation. Our comparison with nine classical and state-of-the-art imputation methods on 24 small to moderate-dimensional tabular data sets shows that CFMI matches or outperforms both traditional and modern techniques across a wide range of metrics. Applying the method to zero-shot imputation of time-series data, we find that it m"},"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":"2506.09258","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-06-10T21:40:36Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"abc9af68d1bf27a5c47062ad127b1e27bfb19ea15a4790fccab1f59ba6fd6d0b","abstract_canon_sha256":"6e0ef8186ff0e4d5f46c1a3dd108bb43dfb2fe288c3b269980b8c74efc71c018"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:19:33.813715Z","signature_b64":"CV6VjhkwaITvTxNmdUmPCWsQMAIZ/zrPqkR2AQw5zQL4LfjIk3vLxWAErF6c+Yes7yO9z8Ta3vFE1N1O+HreAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1decc8e8fa172b52a7e7d68a76173892ea6a1cf5f4cda6ca291136c8c79333c2","last_reissued_at":"2026-07-05T11:19:33.813253Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:19:33.813253Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CFMI: Flow Matching for Missing Data Imputation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Michael U. Gutmann, Vaidotas Simkus","submitted_at":"2025-06-10T21:40:36Z","abstract_excerpt":"We introduce conditional flow matching for imputation (CFMI), a new general-purpose method to impute missing data. The method combines continuous normalising flows, flow-matching, and shared conditional modelling to deal with intractabilities of traditional multiple imputation. Our comparison with nine classical and state-of-the-art imputation methods on 24 small to moderate-dimensional tabular data sets shows that CFMI matches or outperforms both traditional and modern techniques across a wide range of metrics. Applying the method to zero-shot imputation of time-series data, we find that it m"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.09258","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/2506.09258/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":"2506.09258","created_at":"2026-07-05T11:19:33.813310+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.09258v1","created_at":"2026-07-05T11:19:33.813310+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.09258","created_at":"2026-07-05T11:19:33.813310+00:00"},{"alias_kind":"pith_short_12","alias_value":"DXWMR2H2C4VV","created_at":"2026-07-05T11:19:33.813310+00:00"},{"alias_kind":"pith_short_16","alias_value":"DXWMR2H2C4VVFJ7H","created_at":"2026-07-05T11:19:33.813310+00:00"},{"alias_kind":"pith_short_8","alias_value":"DXWMR2H2","created_at":"2026-07-05T11:19:33.813310+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06328","citing_title":"PAMF: Prior-Aware Multimodal Fusion for Incomplete Time Series Data","ref_index":37,"is_internal_anchor":false},{"citing_arxiv_id":"2509.20098","citing_title":"Incomplete Data, Complete Dynamics: A Diffusion Approach","ref_index":40,"is_internal_anchor":false},{"citing_arxiv_id":"2604.18319","citing_title":"Overcoming Selection Bias in Statistical Studies With Amortized Bayesian Inference","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL","json":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL.json","graph_json":"https://pith.science/api/pith-number/DXWMR2H2C4VVFJ7H22FHMFZYSL/graph.json","events_json":"https://pith.science/api/pith-number/DXWMR2H2C4VVFJ7H22FHMFZYSL/events.json","paper":"https://pith.science/paper/DXWMR2H2"},"agent_actions":{"view_html":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL","download_json":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL.json","view_paper":"https://pith.science/paper/DXWMR2H2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.09258&json=true","fetch_graph":"https://pith.science/api/pith-number/DXWMR2H2C4VVFJ7H22FHMFZYSL/graph.json","fetch_events":"https://pith.science/api/pith-number/DXWMR2H2C4VVFJ7H22FHMFZYSL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL/action/storage_attestation","attest_author":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL/action/author_attestation","sign_citation":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL/action/citation_signature","submit_replication":"https://pith.science/pith/DXWMR2H2C4VVFJ7H22FHMFZYSL/action/replication_record"}},"created_at":"2026-07-05T11:19:33.813310+00:00","updated_at":"2026-07-05T11:19:33.813310+00:00"}