{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:EA3JSD277QKMVYOVKH2D56OG2C","short_pith_number":"pith:EA3JSD27","schema_version":"1.0","canonical_sha256":"2036990f5ffc14cae1d551f43ef9c6d084cef00d1a57f6631495cc7d99fdc36e","source":{"kind":"arxiv","id":"2306.08009","version":1},"attestation_state":"computed","paper":{"title":"DHBE: Data-free Holistic Backdoor Erasing in Deep Neural Networks via Restricted Adversarial Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Ruijie Zhao, Shenghong Li, Yuan Tian, Yuanyuan Zhao, Zhicong Yan","submitted_at":"2023-06-13T09:29:05Z","abstract_excerpt":"Backdoor attacks have emerged as an urgent threat to Deep Neural Networks (DNNs), where victim DNNs are furtively implanted with malicious neurons that could be triggered by the adversary. To defend against backdoor attacks, many works establish a staged pipeline to remove backdoors from victim DNNs: inspecting, locating, and erasing. However, in a scenario where a few clean data can be accessible, such pipeline is fragile and cannot erase backdoors completely without sacrificing model accuracy. To address this issue, in this paper, we propose a novel data-free holistic backdoor erasing (DHBE)"},"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":"2306.08009","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-06-13T09:29:05Z","cross_cats_sorted":["cs.AI","cs.CR"],"title_canon_sha256":"5653e2ad7a603bf9b9bf64ce8549ff59ee235b60272149c57513ed2966f6178f","abstract_canon_sha256":"06e5c84970504253862fcd60012e52406d71c0dfbb3f95adcc7b316c9fdaa68d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:20:46.644547Z","signature_b64":"nlEfDK1ybTMVldcfp6lTJy7uEvLK+9bL8N82lUl7g7u5NhGfmq5/LJLvI1I3qe3QJWKsq3eA1rEGromTGAMdDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2036990f5ffc14cae1d551f43ef9c6d084cef00d1a57f6631495cc7d99fdc36e","last_reissued_at":"2026-07-05T06:20:46.644099Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:20:46.644099Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DHBE: Data-free Holistic Backdoor Erasing in Deep Neural Networks via Restricted Adversarial Distillation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CR"],"primary_cat":"cs.LG","authors_text":"Ruijie Zhao, Shenghong Li, Yuan Tian, Yuanyuan Zhao, Zhicong Yan","submitted_at":"2023-06-13T09:29:05Z","abstract_excerpt":"Backdoor attacks have emerged as an urgent threat to Deep Neural Networks (DNNs), where victim DNNs are furtively implanted with malicious neurons that could be triggered by the adversary. To defend against backdoor attacks, many works establish a staged pipeline to remove backdoors from victim DNNs: inspecting, locating, and erasing. However, in a scenario where a few clean data can be accessible, such pipeline is fragile and cannot erase backdoors completely without sacrificing model accuracy. To address this issue, in this paper, we propose a novel data-free holistic backdoor erasing (DHBE)"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.08009","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/2306.08009/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":"2306.08009","created_at":"2026-07-05T06:20:46.644157+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.08009v1","created_at":"2026-07-05T06:20:46.644157+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.08009","created_at":"2026-07-05T06:20:46.644157+00:00"},{"alias_kind":"pith_short_12","alias_value":"EA3JSD277QKM","created_at":"2026-07-05T06:20:46.644157+00:00"},{"alias_kind":"pith_short_16","alias_value":"EA3JSD277QKMVYOV","created_at":"2026-07-05T06:20:46.644157+00:00"},{"alias_kind":"pith_short_8","alias_value":"EA3JSD27","created_at":"2026-07-05T06:20:46.644157+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/EA3JSD277QKMVYOVKH2D56OG2C","json":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C.json","graph_json":"https://pith.science/api/pith-number/EA3JSD277QKMVYOVKH2D56OG2C/graph.json","events_json":"https://pith.science/api/pith-number/EA3JSD277QKMVYOVKH2D56OG2C/events.json","paper":"https://pith.science/paper/EA3JSD27"},"agent_actions":{"view_html":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C","download_json":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C.json","view_paper":"https://pith.science/paper/EA3JSD27","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.08009&json=true","fetch_graph":"https://pith.science/api/pith-number/EA3JSD277QKMVYOVKH2D56OG2C/graph.json","fetch_events":"https://pith.science/api/pith-number/EA3JSD277QKMVYOVKH2D56OG2C/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C/action/storage_attestation","attest_author":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C/action/author_attestation","sign_citation":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C/action/citation_signature","submit_replication":"https://pith.science/pith/EA3JSD277QKMVYOVKH2D56OG2C/action/replication_record"}},"created_at":"2026-07-05T06:20:46.644157+00:00","updated_at":"2026-07-05T06:20:46.644157+00:00"}