{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TKADPU2KVH3FH6GKSPFBC3VGFU","short_pith_number":"pith:TKADPU2K","schema_version":"1.0","canonical_sha256":"9a8037d34aa9f653f8ca93ca116ea62d0d1ae4771e2ccd9ffc525717c5a44e64","source":{"kind":"arxiv","id":"2604.19219","version":2},"attestation_state":"computed","paper":{"title":"Sherpa.ai Privacy-Preserving Multi-Party Entity Alignment without Intersection Disclosure for Noisy Identifiers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"A multi-party private set union protocol enables privacy-preserving entity alignment in vertical federated learning without disclosing shared samples.","cross_cats":["cs.AI","cs.DC","cs.LG"],"primary_cat":"cs.CR","authors_text":"Daniel M. Jimenez-Gutierrez, Dario Pighin, Enrique Zuazua, Georgios Kellaris, Joaquin Del Rio, Oleksii Sliusarenko, Xabi Uribe-Etxebarria","submitted_at":"2026-04-21T08:24:07Z","abstract_excerpt":"Federated Learning (FL) enables collaborative model training among multiple parties without centralizing raw data. There are two main paradigms in FL: Horizontal FL (HFL), where all participants share the same feature space but hold different samples, and Vertical FL (VFL), where parties possess complementary features for the same set of samples. A prerequisite for VFL training is privacy-preserving entity alignment (PPEA), which establishes a common index of samples across parties (alignment) without revealing which samples are shared between them. Conventional private set intersection (PSI) "},"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":"2604.19219","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CR","submitted_at":"2026-04-21T08:24:07Z","cross_cats_sorted":["cs.AI","cs.DC","cs.LG"],"title_canon_sha256":"2bdb4aae4f41a0384103f9f297098a36f2dcbc783ad0f60016c5aee5fd1bc360","abstract_canon_sha256":"b9a1482226f2fc899f5ae1495d512bc32f983d65f025b7bd9b9be9cce8c2f8f0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-20T00:03:12.033709Z","signature_b64":"FtFpdZ7KJ8vnegIyEVO6ol/UJqmP0h3my7V2d6xbcjSU2eiRjg61gDsWqIoOoh/dGowAT95qn9RbliE2L6ynAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9a8037d34aa9f653f8ca93ca116ea62d0d1ae4771e2ccd9ffc525717c5a44e64","last_reissued_at":"2026-05-20T00:03:12.032741Z","signature_status":"signed_v1","first_computed_at":"2026-05-20T00:03:12.032741Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sherpa.ai Privacy-Preserving Multi-Party Entity Alignment without Intersection Disclosure for Noisy Identifiers","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"A multi-party private set union protocol enables privacy-preserving entity alignment in vertical federated learning without disclosing shared samples.","cross_cats":["cs.AI","cs.DC","cs.LG"],"primary_cat":"cs.CR","authors_text":"Daniel M. Jimenez-Gutierrez, Dario Pighin, Enrique Zuazua, Georgios Kellaris, Joaquin Del Rio, Oleksii Sliusarenko, Xabi Uribe-Etxebarria","submitted_at":"2026-04-21T08:24:07Z","abstract_excerpt":"Federated Learning (FL) enables collaborative model training among multiple parties without centralizing raw data. There are two main paradigms in FL: Horizontal FL (HFL), where all participants share the same feature space but hold different samples, and Vertical FL (VFL), where parties possess complementary features for the same set of samples. A prerequisite for VFL training is privacy-preserving entity alignment (PPEA), which establishes a common index of samples across parties (alignment) without revealing which samples are shared between them. Conventional private set intersection (PSI) "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"We introduce the Sherpa.ai multi-party PSU protocol for VFL, a PPEA method that hides intersection membership and enables both exact and noisy matching. We prove correctness and privacy, analyze communication and computational (exponentiation) complexity, and formalize a universal index mapping from local records to a shared index space.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"The protocol assumes that standard cryptographic primitives for multi-party computation can be securely composed to achieve the claimed privacy properties against the stated adversary model, and that the unordered variant can reliably handle the specific types of typographical and formatting noise present in real identifiers without additional false positives or privacy leakage.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"Sherpa.ai presents a multi-party PSU protocol for vertical federated learning that hides intersection membership and supports exact and noisy identifier matching with proven correctness and privacy.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"A multi-party private set union protocol enables privacy-preserving entity alignment in vertical federated learning without disclosing shared samples.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"0bfb7021abc37be066b830bb503e22e1d0595d705588386713702042f98c071e"},"source":{"id":"2604.19219","kind":"arxiv","version":2},"verdict":{"id":"54484485-6c37-465c-834d-19b386fd7d14","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-10T02:56:28.649775Z","strongest_claim":"We introduce the Sherpa.ai multi-party PSU protocol for VFL, a PPEA method that hides intersection membership and enables both exact and noisy matching. We prove correctness and privacy, analyze communication and computational (exponentiation) complexity, and formalize a universal index mapping from local records to a shared index space.","one_line_summary":"Sherpa.ai presents a multi-party PSU protocol for vertical federated learning that hides intersection membership and supports exact and noisy identifier matching with proven correctness and privacy.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"The protocol assumes that standard cryptographic primitives for multi-party computation can be securely composed to achieve the claimed privacy properties against the stated adversary model, and that the unordered variant can reliably handle the specific types of typographical and formatting noise present in real identifiers without additional false positives or privacy leakage.","pith_extraction_headline":"A multi-party private set union protocol enables privacy-preserving entity alignment in vertical federated learning without disclosing shared samples."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.19219/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":"2604.19219","created_at":"2026-05-20T00:03:12.032878+00:00"},{"alias_kind":"arxiv_version","alias_value":"2604.19219v2","created_at":"2026-05-20T00:03:12.032878+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.19219","created_at":"2026-05-20T00:03:12.032878+00:00"},{"alias_kind":"pith_short_12","alias_value":"TKADPU2KVH3F","created_at":"2026-05-20T00:03:12.032878+00:00"},{"alias_kind":"pith_short_16","alias_value":"TKADPU2KVH3FH6GK","created_at":"2026-05-20T00:03:12.032878+00:00"},{"alias_kind":"pith_short_8","alias_value":"TKADPU2K","created_at":"2026-05-20T00:03:12.032878+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/TKADPU2KVH3FH6GKSPFBC3VGFU","json":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU.json","graph_json":"https://pith.science/api/pith-number/TKADPU2KVH3FH6GKSPFBC3VGFU/graph.json","events_json":"https://pith.science/api/pith-number/TKADPU2KVH3FH6GKSPFBC3VGFU/events.json","paper":"https://pith.science/paper/TKADPU2K"},"agent_actions":{"view_html":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU","download_json":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU.json","view_paper":"https://pith.science/paper/TKADPU2K","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2604.19219&json=true","fetch_graph":"https://pith.science/api/pith-number/TKADPU2KVH3FH6GKSPFBC3VGFU/graph.json","fetch_events":"https://pith.science/api/pith-number/TKADPU2KVH3FH6GKSPFBC3VGFU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU/action/storage_attestation","attest_author":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU/action/author_attestation","sign_citation":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU/action/citation_signature","submit_replication":"https://pith.science/pith/TKADPU2KVH3FH6GKSPFBC3VGFU/action/replication_record"}},"created_at":"2026-05-20T00:03:12.032878+00:00","updated_at":"2026-05-20T00:03:12.032878+00:00"}