{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:VWVP2DNVE4D54OBMR767A37T4E","short_pith_number":"pith:VWVP2DNV","schema_version":"1.0","canonical_sha256":"adaafd0db52707de382c8ffdf06ff3e12c039633caf901dba499a5fc2f396eac","source":{"kind":"arxiv","id":"2512.03238","version":2},"attestation_state":"computed","paper":{"title":"How to DP-fy Your Data: A Practical Guide to Generating Synthetic Data With Differential Privacy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Alex Bie, Alex Kurakin, Andreas Terzis, Crist\\'obal Guzm\\'an, Da Yu, Galen Andrew, H. Brendan McMahan, Lucas Rosenblatt, Morteza Zadimoghaddam, Natalia Ponomareva, Peter Kairouz, Ryan McKenna, Sergei Vassilvitskii, Vincent Cohen-Addad, Zheng Xu","submitted_at":"2025-12-02T21:14:39Z","abstract_excerpt":"High quality data is needed to unlock the full potential of AI for end users. However finding new sources of such data is getting harder: most publicly-available human generated data will soon have been used. Additionally, publicly available data often is not representative of users of a particular system -- for example, a research speech dataset of contractors interacting with an AI assistant will likely be more homogeneous, well articulated and self-censored than real world commands that end users will issue. Therefore unlocking high-quality data grounded in real user interactions is of vita"},"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":"2512.03238","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-12-02T21:14:39Z","cross_cats_sorted":["cs.AI","cs.LG","stat.ML"],"title_canon_sha256":"319b31d8cd43267858353bbf59bcce61f0c633bd0d35409c5bd4ccd031e65084","abstract_canon_sha256":"a1b064c2e6b571efe5a5c574b1fae6f491a04f73df1daa46f90bd849e81c2fdc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-13T00:17:04.854278Z","signature_b64":"RW/pklKn2WSPO49MOXj7reWksaWRQmhYrndma5V2/dGlZJXJ7RTcLEM05iX3HPBeWWSTjNKV2MDxKCAEFNQNDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"adaafd0db52707de382c8ffdf06ff3e12c039633caf901dba499a5fc2f396eac","last_reissued_at":"2026-07-13T00:17:04.853036Z","signature_status":"signed_v1","first_computed_at":"2026-07-13T00:17:04.853036Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"How to DP-fy Your Data: A Practical Guide to Generating Synthetic Data With Differential Privacy","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Alex Bie, Alex Kurakin, Andreas Terzis, Crist\\'obal Guzm\\'an, Da Yu, Galen Andrew, H. Brendan McMahan, Lucas Rosenblatt, Morteza Zadimoghaddam, Natalia Ponomareva, Peter Kairouz, Ryan McKenna, Sergei Vassilvitskii, Vincent Cohen-Addad, Zheng Xu","submitted_at":"2025-12-02T21:14:39Z","abstract_excerpt":"High quality data is needed to unlock the full potential of AI for end users. However finding new sources of such data is getting harder: most publicly-available human generated data will soon have been used. Additionally, publicly available data often is not representative of users of a particular system -- for example, a research speech dataset of contractors interacting with an AI assistant will likely be more homogeneous, well articulated and self-censored than real world commands that end users will issue. Therefore unlocking high-quality data grounded in real user interactions is of vita"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2512.03238","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/2512.03238/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":"2512.03238","created_at":"2026-07-13T00:17:04.853613+00:00"},{"alias_kind":"arxiv_version","alias_value":"2512.03238v2","created_at":"2026-07-13T00:17:04.853613+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2512.03238","created_at":"2026-07-13T00:17:04.853613+00:00"},{"alias_kind":"pith_short_12","alias_value":"VWVP2DNVE4D5","created_at":"2026-07-13T00:17:04.853613+00:00"},{"alias_kind":"pith_short_16","alias_value":"VWVP2DNVE4D54OBM","created_at":"2026-07-13T00:17:04.853613+00:00"},{"alias_kind":"pith_short_8","alias_value":"VWVP2DNV","created_at":"2026-07-13T00:17:04.853613+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2606.11283","citing_title":"Fixed-Parameter Tractability of Private Synthetic Data Generation","ref_index":49,"is_internal_anchor":true},{"citing_arxiv_id":"2606.10481","citing_title":"Advancing the State-of-the-Art in Empirical Privacy Auditing","ref_index":75,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01849","citing_title":"ContinuousBench: Can Differentially Private Synthetic Text Improve Capabilities?","ref_index":13,"is_internal_anchor":true},{"citing_arxiv_id":"2605.06211","citing_title":"Contrastive Identification and Generation in the Limit","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E","json":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E.json","graph_json":"https://pith.science/api/pith-number/VWVP2DNVE4D54OBMR767A37T4E/graph.json","events_json":"https://pith.science/api/pith-number/VWVP2DNVE4D54OBMR767A37T4E/events.json","paper":"https://pith.science/paper/VWVP2DNV"},"agent_actions":{"view_html":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E","download_json":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E.json","view_paper":"https://pith.science/paper/VWVP2DNV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2512.03238&json=true","fetch_graph":"https://pith.science/api/pith-number/VWVP2DNVE4D54OBMR767A37T4E/graph.json","fetch_events":"https://pith.science/api/pith-number/VWVP2DNVE4D54OBMR767A37T4E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E/action/storage_attestation","attest_author":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E/action/author_attestation","sign_citation":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E/action/citation_signature","submit_replication":"https://pith.science/pith/VWVP2DNVE4D54OBMR767A37T4E/action/replication_record"}},"created_at":"2026-07-13T00:17:04.853613+00:00","updated_at":"2026-07-13T00:17:04.853613+00:00"}