{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:3YPHP27XOOD45PVX4XJUHWADMT","short_pith_number":"pith:3YPHP27X","schema_version":"1.0","canonical_sha256":"de1e77ebf77387cebeb7e5d343d80364cff6d4d872b0f53b2e0910f2b1aaa334","source":{"kind":"arxiv","id":"1801.07698","version":4},"attestation_state":"computed","paper":{"title":"ArcFace: Additive Angular Margin Loss for Deep Face Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Irene Kotsia, Jia Guo, Jiankang Deng, Jing Yang, Niannan Xue, Stefanos Zafeiriou","submitted_at":"2018-01-23T18:39:19Z","abstract_excerpt":"Recently, a popular line of research in face recognition is adopting margins in the well-established softmax loss function to maximize class separability. In this paper, we first introduce an Additive Angular Margin Loss (ArcFace), which not only has a clear geometric interpretation but also significantly enhances the discriminative power. Since ArcFace is susceptible to the massive label noise, we further propose sub-center ArcFace, in which each class contains $K$ sub-centers and training samples only need to be close to any of the $K$ positive sub-centers. Sub-center ArcFace encourages one "},"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":"1801.07698","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2018-01-23T18:39:19Z","cross_cats_sorted":[],"title_canon_sha256":"55efc9150e3ec4caec2be63ec8a0b086caa6848d86d394fdf83b36823227c49b","abstract_canon_sha256":"e2091205d3cc1b6e0bab556df805375f2186aea262fa8a065a9f329a85cde450"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:54:10.670428Z","signature_b64":"nOMVzlnpYTgThMOSj7+GKODijmxDZQBukOI5Et4KPsZU8AQN6gIe7W604UMJ092gu5gleBcc6b+a9Wx7dMpvAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de1e77ebf77387cebeb7e5d343d80364cff6d4d872b0f53b2e0910f2b1aaa334","last_reissued_at":"2026-07-05T04:54:10.669862Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:54:10.669862Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ArcFace: Additive Angular Margin Loss for Deep Face Recognition","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Irene Kotsia, Jia Guo, Jiankang Deng, Jing Yang, Niannan Xue, Stefanos Zafeiriou","submitted_at":"2018-01-23T18:39:19Z","abstract_excerpt":"Recently, a popular line of research in face recognition is adopting margins in the well-established softmax loss function to maximize class separability. In this paper, we first introduce an Additive Angular Margin Loss (ArcFace), which not only has a clear geometric interpretation but also significantly enhances the discriminative power. Since ArcFace is susceptible to the massive label noise, we further propose sub-center ArcFace, in which each class contains $K$ sub-centers and training samples only need to be close to any of the $K$ positive sub-centers. Sub-center ArcFace encourages one "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1801.07698","kind":"arxiv","version":4},"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/1801.07698/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":"1801.07698","created_at":"2026-07-05T04:54:10.669930+00:00"},{"alias_kind":"arxiv_version","alias_value":"1801.07698v4","created_at":"2026-07-05T04:54:10.669930+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1801.07698","created_at":"2026-07-05T04:54:10.669930+00:00"},{"alias_kind":"pith_short_12","alias_value":"3YPHP27XOOD4","created_at":"2026-07-05T04:54:10.669930+00:00"},{"alias_kind":"pith_short_16","alias_value":"3YPHP27XOOD45PVX","created_at":"2026-07-05T04:54:10.669930+00:00"},{"alias_kind":"pith_short_8","alias_value":"3YPHP27X","created_at":"2026-07-05T04:54:10.669930+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":8,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.04469","citing_title":"Adaptive Calibration for Fair and Performant Facial Recognition","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"1906.08507","citing_title":"Multiple-Identity Image Attacks Against Face-based Identity Verification","ref_index":63,"is_internal_anchor":false},{"citing_arxiv_id":"1906.10546","citing_title":"Knowledge Amalgamation from Heterogeneous Networks by Common Feature Learning","ref_index":2,"is_internal_anchor":false},{"citing_arxiv_id":"1907.02642","citing_title":"Primate Face Identification in the Wild","ref_index":8,"is_internal_anchor":false},{"citing_arxiv_id":"1907.10473","citing_title":"Switchable Normalization for Learning-to-Normalize Deep Representation","ref_index":55,"is_internal_anchor":false},{"citing_arxiv_id":"2605.19032","citing_title":"Personalized Face Privacy Protection From a Single Image","ref_index":7,"is_internal_anchor":false},{"citing_arxiv_id":"2603.26571","citing_title":"GVCC: Zero-Shot Video Compression via Codebook-Driven Stochastic Rectified Flow","ref_index":10,"is_internal_anchor":false},{"citing_arxiv_id":"2604.14734","citing_title":"Find the Differences: Differential Morphing Attack Detection vs Face Recognition","ref_index":24,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT","json":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT.json","graph_json":"https://pith.science/api/pith-number/3YPHP27XOOD45PVX4XJUHWADMT/graph.json","events_json":"https://pith.science/api/pith-number/3YPHP27XOOD45PVX4XJUHWADMT/events.json","paper":"https://pith.science/paper/3YPHP27X"},"agent_actions":{"view_html":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT","download_json":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT.json","view_paper":"https://pith.science/paper/3YPHP27X","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1801.07698&json=true","fetch_graph":"https://pith.science/api/pith-number/3YPHP27XOOD45PVX4XJUHWADMT/graph.json","fetch_events":"https://pith.science/api/pith-number/3YPHP27XOOD45PVX4XJUHWADMT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT/action/storage_attestation","attest_author":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT/action/author_attestation","sign_citation":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT/action/citation_signature","submit_replication":"https://pith.science/pith/3YPHP27XOOD45PVX4XJUHWADMT/action/replication_record"}},"created_at":"2026-07-05T04:54:10.669930+00:00","updated_at":"2026-07-05T04:54:10.669930+00:00"}