{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:45SHVBPIKDLQFCVYJEECJMNOKD","short_pith_number":"pith:45SHVBPI","schema_version":"1.0","canonical_sha256":"e7647a85e850d7028ab8490824b1ae50c4533f3962ab7844b051ab9fe4b86da9","source":{"kind":"arxiv","id":"2505.18380","version":2},"attestation_state":"computed","paper":{"title":"RedactOR: An LLM-Powered Framework for Automatic Clinical Data De-Identification","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Charlotte Dzialo, Irfan Bulu, Jangwon Kim, Krishnaram Kenthapadi, Praphul Singh, Sri Gadde, Sumana Srivatsa","submitted_at":"2025-05-23T21:13:18Z","abstract_excerpt":"Ensuring clinical data privacy while preserving utility is critical for AI-driven healthcare and data analytics. Existing de-identification (De-ID) methods, including rule-based techniques, deep learning models, and large language models (LLMs), often suffer from recall errors, limited generalization, and inefficiencies, limiting their real-world applicability. We propose a fully automated, multi-modal framework, RedactOR for de-identifying structured and unstructured electronic health records, including clinical audio records. Our framework employs cost-efficient De-ID strategies, including i"},"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":"2505.18380","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2025-05-23T21:13:18Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"6fe3b6a9881ff470589790dff1cc899918c1b12245c9721ed0981b3177923539","abstract_canon_sha256":"0a0153400c944d5bf1e110117c455d7a549cdb9e408454e990089348c6db038b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:42:54.529421Z","signature_b64":"f6eqlylm0AokkQGlrpdpQUw+B0+DNNykGK+4Jm1S3/vSnpCqf6fY64eAOp3zBpQHMp4vry45Q3HpwGysI4lSDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7647a85e850d7028ab8490824b1ae50c4533f3962ab7844b051ab9fe4b86da9","last_reissued_at":"2026-07-05T11:42:54.528934Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:42:54.528934Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RedactOR: An LLM-Powered Framework for Automatic Clinical Data De-Identification","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.AI","authors_text":"Charlotte Dzialo, Irfan Bulu, Jangwon Kim, Krishnaram Kenthapadi, Praphul Singh, Sri Gadde, Sumana Srivatsa","submitted_at":"2025-05-23T21:13:18Z","abstract_excerpt":"Ensuring clinical data privacy while preserving utility is critical for AI-driven healthcare and data analytics. Existing de-identification (De-ID) methods, including rule-based techniques, deep learning models, and large language models (LLMs), often suffer from recall errors, limited generalization, and inefficiencies, limiting their real-world applicability. We propose a fully automated, multi-modal framework, RedactOR for de-identifying structured and unstructured electronic health records, including clinical audio records. Our framework employs cost-efficient De-ID strategies, including i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.18380","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/2505.18380/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":"2505.18380","created_at":"2026-07-05T11:42:54.528987+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.18380v2","created_at":"2026-07-05T11:42:54.528987+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.18380","created_at":"2026-07-05T11:42:54.528987+00:00"},{"alias_kind":"pith_short_12","alias_value":"45SHVBPIKDLQ","created_at":"2026-07-05T11:42:54.528987+00:00"},{"alias_kind":"pith_short_16","alias_value":"45SHVBPIKDLQFCVY","created_at":"2026-07-05T11:42:54.528987+00:00"},{"alias_kind":"pith_short_8","alias_value":"45SHVBPI","created_at":"2026-07-05T11:42:54.528987+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.03301","citing_title":"SHIELD: A Diverse Clinical Note Dataset and Distilled Small Language Models for Enterprise-Scale De-identification","ref_index":32,"is_internal_anchor":false},{"citing_arxiv_id":"2605.03301","citing_title":"SHIELD: A Diverse Clinical Note Dataset and Distilled Small Language Models for Enterprise-Scale De-identification","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD","json":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD.json","graph_json":"https://pith.science/api/pith-number/45SHVBPIKDLQFCVYJEECJMNOKD/graph.json","events_json":"https://pith.science/api/pith-number/45SHVBPIKDLQFCVYJEECJMNOKD/events.json","paper":"https://pith.science/paper/45SHVBPI"},"agent_actions":{"view_html":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD","download_json":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD.json","view_paper":"https://pith.science/paper/45SHVBPI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.18380&json=true","fetch_graph":"https://pith.science/api/pith-number/45SHVBPIKDLQFCVYJEECJMNOKD/graph.json","fetch_events":"https://pith.science/api/pith-number/45SHVBPIKDLQFCVYJEECJMNOKD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD/action/storage_attestation","attest_author":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD/action/author_attestation","sign_citation":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD/action/citation_signature","submit_replication":"https://pith.science/pith/45SHVBPIKDLQFCVYJEECJMNOKD/action/replication_record"}},"created_at":"2026-07-05T11:42:54.528987+00:00","updated_at":"2026-07-05T11:42:54.528987+00:00"}