{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:5CSAC4UIZQ3LGSHX2736WGVZT3","short_pith_number":"pith:5CSAC4UI","schema_version":"1.0","canonical_sha256":"e8a4017288cc36b348f7d7f7eb1ab99ef86231a91f1b31fa0c50b1950db64661","source":{"kind":"arxiv","id":"2301.03118","version":2},"attestation_state":"computed","paper":{"title":"Facial Misrecognition Systems: Simple Weight Manipulations Force DNNs to Err Only on Specific Persons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Adi Shamir, Irad Zehavi, Roee Nitzan","submitted_at":"2023-01-08T23:11:00Z","abstract_excerpt":"In this paper, we describe how to plant novel types of backdoors in any facial recognition model based on the popular architecture of deep Siamese neural networks. These backdoors force the system to err only on natural images of specific persons who are preselected by the attacker, without controlling their appearance or inserting any triggers. For example, we show how such a backdoored system can classify any two images of a particular person as different people, or any two images of a particular pair of persons as the same person, with almost no effect on the correctness of its decisions fo"},"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":"2301.03118","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CR","submitted_at":"2023-01-08T23:11:00Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1a8cca98ba59cdeb608d4f5243d9d99acab925bb6797abff0e3ba947ff5aa94a","abstract_canon_sha256":"1989434a4f3ddb8770d8d0ef2ac1744b2748f35b01403abad61c2958ecb1efa8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:30:28.304721Z","signature_b64":"qJDNLXek4qrEDr2QAT8WUdOPYyLv8/Zp75rMQXb9smsHSM+lOj3Iqp8w9NnGLGVRm7JEwcq1oi1PGGpHPnevBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e8a4017288cc36b348f7d7f7eb1ab99ef86231a91f1b31fa0c50b1950db64661","last_reissued_at":"2026-07-05T08:30:28.304279Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:30:28.304279Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Facial Misrecognition Systems: Simple Weight Manipulations Force DNNs to Err Only on Specific Persons","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CR","authors_text":"Adi Shamir, Irad Zehavi, Roee Nitzan","submitted_at":"2023-01-08T23:11:00Z","abstract_excerpt":"In this paper, we describe how to plant novel types of backdoors in any facial recognition model based on the popular architecture of deep Siamese neural networks. These backdoors force the system to err only on natural images of specific persons who are preselected by the attacker, without controlling their appearance or inserting any triggers. For example, we show how such a backdoored system can classify any two images of a particular person as different people, or any two images of a particular pair of persons as the same person, with almost no effect on the correctness of its decisions fo"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.03118","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/2301.03118/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":"2301.03118","created_at":"2026-07-05T08:30:28.304349+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.03118v2","created_at":"2026-07-05T08:30:28.304349+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.03118","created_at":"2026-07-05T08:30:28.304349+00:00"},{"alias_kind":"pith_short_12","alias_value":"5CSAC4UIZQ3L","created_at":"2026-07-05T08:30:28.304349+00:00"},{"alias_kind":"pith_short_16","alias_value":"5CSAC4UIZQ3LGSHX","created_at":"2026-07-05T08:30:28.304349+00:00"},{"alias_kind":"pith_short_8","alias_value":"5CSAC4UI","created_at":"2026-07-05T08:30:28.304349+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.20307","citing_title":"Surveying the Operational Cybersecurity and Supply Chain Threat Landscape when Developing and Deploying AI Systems","ref_index":39,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3","json":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3.json","graph_json":"https://pith.science/api/pith-number/5CSAC4UIZQ3LGSHX2736WGVZT3/graph.json","events_json":"https://pith.science/api/pith-number/5CSAC4UIZQ3LGSHX2736WGVZT3/events.json","paper":"https://pith.science/paper/5CSAC4UI"},"agent_actions":{"view_html":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3","download_json":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3.json","view_paper":"https://pith.science/paper/5CSAC4UI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.03118&json=true","fetch_graph":"https://pith.science/api/pith-number/5CSAC4UIZQ3LGSHX2736WGVZT3/graph.json","fetch_events":"https://pith.science/api/pith-number/5CSAC4UIZQ3LGSHX2736WGVZT3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3/action/storage_attestation","attest_author":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3/action/author_attestation","sign_citation":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3/action/citation_signature","submit_replication":"https://pith.science/pith/5CSAC4UIZQ3LGSHX2736WGVZT3/action/replication_record"}},"created_at":"2026-07-05T08:30:28.304349+00:00","updated_at":"2026-07-05T08:30:28.304349+00:00"}