{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HIGU6KMJPBNL5SURXTT2YNJTCA","short_pith_number":"pith:HIGU6KMJ","schema_version":"1.0","canonical_sha256":"3a0d4f2989785abeca91bce7ac3533100c1a5cd966a97b6535365ebfb07089e5","source":{"kind":"arxiv","id":"2108.04978","version":1},"attestation_state":"computed","paper":{"title":"Winning the NIST Contest: A scalable and general approach to differentially private synthetic data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Daniel Sheldon, Gerome Miklau, Ryan McKenna","submitted_at":"2021-08-11T00:49:48Z","abstract_excerpt":"We propose a general approach for differentially private synthetic data generation, that consists of three steps: (1) select a collection of low-dimensional marginals, (2) measure those marginals with a noise addition mechanism, and (3) generate synthetic data that preserves the measured marginals well. Central to this approach is Private-PGM, a post-processing method that is used to estimate a high-dimensional data distribution from noisy measurements of its marginals. We present two mechanisms, NIST-MST and MST, that are instances of this general approach. NIST-MST was the winning mechanism "},"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":"2108.04978","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2021-08-11T00:49:48Z","cross_cats_sorted":[],"title_canon_sha256":"256b94ee69d635a815ffd289957dd0191c6a4a867bb57f0d68baf2a72b9ac61d","abstract_canon_sha256":"26ff36716f2a9ee33c027e3d34f7d6d251caf57f9d93bac13d8a80105fafae1c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:05:05.696468Z","signature_b64":"/yTMTrTlM9Yc7sAhigueOh8Z79+ZQao8qgl7bc8qsJtBrODoEeuc7/fNmWkIeupHMP5mauX6OgwSDuCRRty5CQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3a0d4f2989785abeca91bce7ac3533100c1a5cd966a97b6535365ebfb07089e5","last_reissued_at":"2026-07-05T03:05:05.695962Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:05:05.695962Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Winning the NIST Contest: A scalable and general approach to differentially private synthetic data","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Daniel Sheldon, Gerome Miklau, Ryan McKenna","submitted_at":"2021-08-11T00:49:48Z","abstract_excerpt":"We propose a general approach for differentially private synthetic data generation, that consists of three steps: (1) select a collection of low-dimensional marginals, (2) measure those marginals with a noise addition mechanism, and (3) generate synthetic data that preserves the measured marginals well. Central to this approach is Private-PGM, a post-processing method that is used to estimate a high-dimensional data distribution from noisy measurements of its marginals. We present two mechanisms, NIST-MST and MST, that are instances of this general approach. NIST-MST was the winning mechanism "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2108.04978","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/2108.04978/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":"2108.04978","created_at":"2026-07-05T03:05:05.696028+00:00"},{"alias_kind":"arxiv_version","alias_value":"2108.04978v1","created_at":"2026-07-05T03:05:05.696028+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2108.04978","created_at":"2026-07-05T03:05:05.696028+00:00"},{"alias_kind":"pith_short_12","alias_value":"HIGU6KMJPBNL","created_at":"2026-07-05T03:05:05.696028+00:00"},{"alias_kind":"pith_short_16","alias_value":"HIGU6KMJPBNL5SUR","created_at":"2026-07-05T03:05:05.696028+00:00"},{"alias_kind":"pith_short_8","alias_value":"HIGU6KMJ","created_at":"2026-07-05T03:05:05.696028+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2607.08122","citing_title":"Workload-Preserving Differentially Private Synthetic Data for Causal Inference via Maximum-Entropy Calibration","ref_index":6,"is_internal_anchor":true},{"citing_arxiv_id":"2607.07471","citing_title":"Where to Intervene? Benchmarking Fairness-Aware Learning on Differentially Private Synthetic Tabular Data","ref_index":46,"is_internal_anchor":true},{"citing_arxiv_id":"2606.08372","citing_title":"SoK: Reconstruction Attacks on Synthetic Tabular Data (Insights from Winning the NIST CRC)","ref_index":46,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06334","citing_title":"Quantifying the Privacy of Counterfactuals by Leveraging Membership Inference Attacks Against Synthetic Data","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2605.22952","citing_title":"Measuring Database Unfairness via Dependency Quantification Under Differential Privacy","ref_index":73,"is_internal_anchor":false},{"citing_arxiv_id":"2507.02971","citing_title":"Aim High, Stay Private: Differentially Private Synthetic Data Enables Public Release of Behavioral Health Information with High Utility","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2604.19653","citing_title":"A Dual Perspective on Synthetic Trajectory Generators: Utility Framework and Privacy Vulnerabilities","ref_index":85,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14495","citing_title":"Decoupling Identity from Utility: Privacy-by-Design Frameworks for Financial Ecosystems","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15660","citing_title":"DPDSyn: Improving Differentially Private Dataset Synthesis for Model Training by Downstream Task Guidance","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA","json":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA.json","graph_json":"https://pith.science/api/pith-number/HIGU6KMJPBNL5SURXTT2YNJTCA/graph.json","events_json":"https://pith.science/api/pith-number/HIGU6KMJPBNL5SURXTT2YNJTCA/events.json","paper":"https://pith.science/paper/HIGU6KMJ"},"agent_actions":{"view_html":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA","download_json":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA.json","view_paper":"https://pith.science/paper/HIGU6KMJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2108.04978&json=true","fetch_graph":"https://pith.science/api/pith-number/HIGU6KMJPBNL5SURXTT2YNJTCA/graph.json","fetch_events":"https://pith.science/api/pith-number/HIGU6KMJPBNL5SURXTT2YNJTCA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA/action/storage_attestation","attest_author":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA/action/author_attestation","sign_citation":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA/action/citation_signature","submit_replication":"https://pith.science/pith/HIGU6KMJPBNL5SURXTT2YNJTCA/action/replication_record"}},"created_at":"2026-07-05T03:05:05.696028+00:00","updated_at":"2026-07-05T03:05:05.696028+00:00"}