{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:O5KK7X543GNUSTH3MBEDK7DEFP","short_pith_number":"pith:O5KK7X54","schema_version":"1.0","canonical_sha256":"7754afdfbcd99b494cfb6048357c642bdae07ac80051d4014105df5e57ae44ed","source":{"kind":"arxiv","id":"2412.05533","version":1},"attestation_state":"computed","paper":{"title":"Can large language models be privacy preserving and fair medical coders?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Ali Dadsetan, Dorsa Soleymani, Frank Rudzicz, Xijie Zeng","submitted_at":"2024-12-07T04:27:05Z","abstract_excerpt":"Protecting patient data privacy is a critical concern when deploying machine learning algorithms in healthcare. Differential privacy (DP) is a common method for preserving privacy in such settings and, in this work, we examine two key trade-offs in applying DP to the NLP task of medical coding (ICD classification). Regarding the privacy-utility trade-off, we observe a significant performance drop in the privacy preserving models, with more than a 40% reduction in micro F1 scores on the top 50 labels in the MIMIC-III dataset. From the perspective of the privacy-fairness trade-off, we also obser"},"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":"2412.05533","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-07T04:27:05Z","cross_cats_sorted":["cs.CR"],"title_canon_sha256":"f1057847ba381d79aba73d81719fd1b477b73924092cfefd7261ec245b515018","abstract_canon_sha256":"62c26206075dc79a95f865ce13c90172b5e6eae4c5cd25722510786d640fb5a3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:46:04.838848Z","signature_b64":"Uyyh5O1BV3YPbRCx6ttKpSX3eEO2S+fo+CA9XU4gKWlNAKHOgek8DyMGMka4dIR/7wYAa1QlGgnsX+K9NfRmBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7754afdfbcd99b494cfb6048357c642bdae07ac80051d4014105df5e57ae44ed","last_reissued_at":"2026-07-05T09:46:04.838446Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:46:04.838446Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can large language models be privacy preserving and fair medical coders?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CR"],"primary_cat":"cs.LG","authors_text":"Ali Dadsetan, Dorsa Soleymani, Frank Rudzicz, Xijie Zeng","submitted_at":"2024-12-07T04:27:05Z","abstract_excerpt":"Protecting patient data privacy is a critical concern when deploying machine learning algorithms in healthcare. Differential privacy (DP) is a common method for preserving privacy in such settings and, in this work, we examine two key trade-offs in applying DP to the NLP task of medical coding (ICD classification). Regarding the privacy-utility trade-off, we observe a significant performance drop in the privacy preserving models, with more than a 40% reduction in micro F1 scores on the top 50 labels in the MIMIC-III dataset. From the perspective of the privacy-fairness trade-off, we also obser"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.05533","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/2412.05533/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":"2412.05533","created_at":"2026-07-05T09:46:04.838502+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.05533v1","created_at":"2026-07-05T09:46:04.838502+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.05533","created_at":"2026-07-05T09:46:04.838502+00:00"},{"alias_kind":"pith_short_12","alias_value":"O5KK7X543GNU","created_at":"2026-07-05T09:46:04.838502+00:00"},{"alias_kind":"pith_short_16","alias_value":"O5KK7X543GNUSTH3","created_at":"2026-07-05T09:46:04.838502+00:00"},{"alias_kind":"pith_short_8","alias_value":"O5KK7X54","created_at":"2026-07-05T09:46:04.838502+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP","json":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP.json","graph_json":"https://pith.science/api/pith-number/O5KK7X543GNUSTH3MBEDK7DEFP/graph.json","events_json":"https://pith.science/api/pith-number/O5KK7X543GNUSTH3MBEDK7DEFP/events.json","paper":"https://pith.science/paper/O5KK7X54"},"agent_actions":{"view_html":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP","download_json":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP.json","view_paper":"https://pith.science/paper/O5KK7X54","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.05533&json=true","fetch_graph":"https://pith.science/api/pith-number/O5KK7X543GNUSTH3MBEDK7DEFP/graph.json","fetch_events":"https://pith.science/api/pith-number/O5KK7X543GNUSTH3MBEDK7DEFP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP/action/storage_attestation","attest_author":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP/action/author_attestation","sign_citation":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP/action/citation_signature","submit_replication":"https://pith.science/pith/O5KK7X543GNUSTH3MBEDK7DEFP/action/replication_record"}},"created_at":"2026-07-05T09:46:04.838502+00:00","updated_at":"2026-07-05T09:46:04.838502+00:00"}