{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:7TNJWVVEEQQX2KDTCWSRZCDNKQ","short_pith_number":"pith:7TNJWVVE","schema_version":"1.0","canonical_sha256":"fcda9b56a424217d287315a51c886d5436313dda5974eea294096b36fb585da4","source":{"kind":"arxiv","id":"2506.15112","version":1},"attestation_state":"computed","paper":{"title":"PDLRecover: Privacy-preserving Decentralized Model Recovery with Machine Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Jianbing Ni, Maazen Alsabaan, Mohamed Mahmoud, Xiangman Li, Xiaodong Wu","submitted_at":"2025-06-18T03:30:07Z","abstract_excerpt":"Decentralized learning is vulnerable to poison attacks, where malicious clients manipulate local updates to degrade global model performance. Existing defenses mainly detect and filter malicious models, aiming to prevent a limited number of attackers from corrupting the global model. However, restoring an already compromised global model remains a challenge. A direct approach is to remove malicious clients and retrain the model using only the benign clients. Yet, retraining is time-consuming, computationally expensive, and may compromise model consistency and privacy.\n  We propose PDLRecover, "},"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":"2506.15112","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CR","submitted_at":"2025-06-18T03:30:07Z","cross_cats_sorted":[],"title_canon_sha256":"f4b0f1c37c477c8d38c086ddc673e5bc926181db1994b47051c29f10f294a61c","abstract_canon_sha256":"6b6151083b5ca9ec077557604263acaaf187b176158d3ee0f037c4e17a1d41ab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:37.281770Z","signature_b64":"rsFN0FTapUUBHUviaOuY0HPx2OSzVxtvwsXcK7IhrhydLII/8qtjBcJckoE6jDrufTGPFEvjIwVAdnNrfcUTBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"fcda9b56a424217d287315a51c886d5436313dda5974eea294096b36fb585da4","last_reissued_at":"2026-07-05T11:23:37.281254Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:37.281254Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"PDLRecover: Privacy-preserving Decentralized Model Recovery with Machine Unlearning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CR","authors_text":"Jianbing Ni, Maazen Alsabaan, Mohamed Mahmoud, Xiangman Li, Xiaodong Wu","submitted_at":"2025-06-18T03:30:07Z","abstract_excerpt":"Decentralized learning is vulnerable to poison attacks, where malicious clients manipulate local updates to degrade global model performance. Existing defenses mainly detect and filter malicious models, aiming to prevent a limited number of attackers from corrupting the global model. However, restoring an already compromised global model remains a challenge. A direct approach is to remove malicious clients and retrain the model using only the benign clients. Yet, retraining is time-consuming, computationally expensive, and may compromise model consistency and privacy.\n  We propose PDLRecover, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.15112","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/2506.15112/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":"2506.15112","created_at":"2026-07-05T11:23:37.281312+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.15112v1","created_at":"2026-07-05T11:23:37.281312+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.15112","created_at":"2026-07-05T11:23:37.281312+00:00"},{"alias_kind":"pith_short_12","alias_value":"7TNJWVVEEQQX","created_at":"2026-07-05T11:23:37.281312+00:00"},{"alias_kind":"pith_short_16","alias_value":"7TNJWVVEEQQX2KDT","created_at":"2026-07-05T11:23:37.281312+00:00"},{"alias_kind":"pith_short_8","alias_value":"7TNJWVVE","created_at":"2026-07-05T11:23:37.281312+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/7TNJWVVEEQQX2KDTCWSRZCDNKQ","json":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ.json","graph_json":"https://pith.science/api/pith-number/7TNJWVVEEQQX2KDTCWSRZCDNKQ/graph.json","events_json":"https://pith.science/api/pith-number/7TNJWVVEEQQX2KDTCWSRZCDNKQ/events.json","paper":"https://pith.science/paper/7TNJWVVE"},"agent_actions":{"view_html":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ","download_json":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ.json","view_paper":"https://pith.science/paper/7TNJWVVE","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.15112&json=true","fetch_graph":"https://pith.science/api/pith-number/7TNJWVVEEQQX2KDTCWSRZCDNKQ/graph.json","fetch_events":"https://pith.science/api/pith-number/7TNJWVVEEQQX2KDTCWSRZCDNKQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ/action/storage_attestation","attest_author":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ/action/author_attestation","sign_citation":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ/action/citation_signature","submit_replication":"https://pith.science/pith/7TNJWVVEEQQX2KDTCWSRZCDNKQ/action/replication_record"}},"created_at":"2026-07-05T11:23:37.281312+00:00","updated_at":"2026-07-05T11:23:37.281312+00:00"}