{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:633HIQDXT3C62NACJT4RJIW4XD","short_pith_number":"pith:633HIQDX","schema_version":"1.0","canonical_sha256":"f6f67440779ec5ed34024cf914a2dcb8e5060d18c4f7abcac7f20d0462af428a","source":{"kind":"arxiv","id":"1811.00513","version":2},"attestation_state":"computed","paper":{"title":"Auditing Data Provenance in Text-Generation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Congzheng Song, Vitaly Shmatikov","submitted_at":"2018-11-01T17:32:44Z","abstract_excerpt":"To help enforce data-protection regulations such as GDPR and detect unauthorized uses of personal data, we develop a new \\emph{model auditing} technique that helps users check if their data was used to train a machine learning model. We focus on auditing deep-learning models that generate natural-language text, including word prediction and dialog generation. These models are at the core of popular online services and are often trained on personal data such as users' messages, searches, chats, and comments.\n  We design and evaluate a black-box auditing method that can detect, with very few que"},"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":"1811.00513","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2018-11-01T17:32:44Z","cross_cats_sorted":["cs.CL","cs.LG","stat.ML"],"title_canon_sha256":"fca88e1826fa0e065b883a36e05d327fed5e5496b94a9f282f2c7f54a5758b91","abstract_canon_sha256":"62c5252e8f62110ea969f4aa169e653baecf41368d113073a48bd260d8b9943e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:45:52.641771Z","signature_b64":"mLxh4ztTcNjWiwDs6EwXOvuevbCNszkfA44ITRUk8zNIRPX+j5keEOIWsK3gKDeFVNmh0wtJNB+wkM6yLXamAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f6f67440779ec5ed34024cf914a2dcb8e5060d18c4f7abcac7f20d0462af428a","last_reissued_at":"2026-05-17T23:45:52.641158Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:45:52.641158Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Auditing Data Provenance in Text-Generation Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.LG","stat.ML"],"primary_cat":"cs.CR","authors_text":"Congzheng Song, Vitaly Shmatikov","submitted_at":"2018-11-01T17:32:44Z","abstract_excerpt":"To help enforce data-protection regulations such as GDPR and detect unauthorized uses of personal data, we develop a new \\emph{model auditing} technique that helps users check if their data was used to train a machine learning model. We focus on auditing deep-learning models that generate natural-language text, including word prediction and dialog generation. These models are at the core of popular online services and are often trained on personal data such as users' messages, searches, chats, and comments.\n  We design and evaluate a black-box auditing method that can detect, with very few que"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.00513","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":""},"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":"1811.00513","created_at":"2026-05-17T23:45:52.641247+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.00513v2","created_at":"2026-05-17T23:45:52.641247+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.00513","created_at":"2026-05-17T23:45:52.641247+00:00"},{"alias_kind":"pith_short_12","alias_value":"633HIQDXT3C6","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"633HIQDXT3C62NAC","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"633HIQDX","created_at":"2026-05-18T12:32:08.215937+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2411.15831","citing_title":"Efficient and Private: Memorisation under differentially private parameter-efficient fine-tuning in language models","ref_index":4,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD","json":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD.json","graph_json":"https://pith.science/api/pith-number/633HIQDXT3C62NACJT4RJIW4XD/graph.json","events_json":"https://pith.science/api/pith-number/633HIQDXT3C62NACJT4RJIW4XD/events.json","paper":"https://pith.science/paper/633HIQDX"},"agent_actions":{"view_html":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD","download_json":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD.json","view_paper":"https://pith.science/paper/633HIQDX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.00513&json=true","fetch_graph":"https://pith.science/api/pith-number/633HIQDXT3C62NACJT4RJIW4XD/graph.json","fetch_events":"https://pith.science/api/pith-number/633HIQDXT3C62NACJT4RJIW4XD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD/action/storage_attestation","attest_author":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD/action/author_attestation","sign_citation":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD/action/citation_signature","submit_replication":"https://pith.science/pith/633HIQDXT3C62NACJT4RJIW4XD/action/replication_record"}},"created_at":"2026-05-17T23:45:52.641247+00:00","updated_at":"2026-05-17T23:45:52.641247+00:00"}