{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:WTYXWUKBKFAEWZOEYX6WJB4BV5","short_pith_number":"pith:WTYXWUKB","canonical_record":{"source":{"id":"1903.10357","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-03-25T14:16:06Z","cross_cats_sorted":[],"title_canon_sha256":"56a85dfedc30952587f0bce3733662a5b76cd37ec70c9cf7c2bb1b4e78bc2dbc","abstract_canon_sha256":"534d1180e9fb219ca40fe2a6493da72e4aa1dc39e2b67fef8a7518471eb42678"},"schema_version":"1.0"},"canonical_sha256":"b4f17b514151404b65c4c5fd648781af54e2f387470a8c5700c57ca1c6b5860b","source":{"kind":"arxiv","id":"1903.10357","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.10357","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"arxiv_version","alias_value":"1903.10357v2","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.10357","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"pith_short_12","alias_value":"WTYXWUKBKFAE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"WTYXWUKBKFAEWZOE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"WTYXWUKB","created_at":"2026-05-18T12:33:30Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:WTYXWUKBKFAEWZOEYX6WJB4BV5","target":"record","payload":{"canonical_record":{"source":{"id":"1903.10357","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-03-25T14:16:06Z","cross_cats_sorted":[],"title_canon_sha256":"56a85dfedc30952587f0bce3733662a5b76cd37ec70c9cf7c2bb1b4e78bc2dbc","abstract_canon_sha256":"534d1180e9fb219ca40fe2a6493da72e4aa1dc39e2b67fef8a7518471eb42678"},"schema_version":"1.0"},"canonical_sha256":"b4f17b514151404b65c4c5fd648781af54e2f387470a8c5700c57ca1c6b5860b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:50:18.627269Z","signature_b64":"8MFq+tfvJfHMDWRT21AKpZG71R8/QyGvn53+hw6STlmZY3X1icdTZ7cqNgQZlEq6wtp5mhE+gUzv0zOl2sm+DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b4f17b514151404b65c4c5fd648781af54e2f387470a8c5700c57ca1c6b5860b","last_reissued_at":"2026-05-17T23:50:18.626642Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:50:18.626642Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1903.10357","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:50:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"JdmS9xZBJDkaA/Y2pFq42dZnO3/ted2YXyhT1k9C5LhcuwnzrBV6QCPDk/Qx3Me+rwucAP674ww+j+oM005CBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T18:22:22.758447Z"},"content_sha256":"0d4c9db1690c3868151672d0d913d94a5f8a3102c0a53350b8cb69ed00d6079f","schema_version":"1.0","event_id":"sha256:0d4c9db1690c3868151672d0d913d94a5f8a3102c0a53350b8cb69ed00d6079f"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:WTYXWUKBKFAEWZOEYX6WJB4BV5","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Noise-Tolerant Paradigm for Training Face Recognition CNNs","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Fan Zhang, Ruirui Li, Wei Hu, Yangyu Huang","submitted_at":"2019-03-25T14:16:06Z","abstract_excerpt":"Benefit from large-scale training datasets, deep Convolutional Neural Networks(CNNs) have achieved impressive results in face recognition(FR). However, tremendous scale of datasets inevitably lead to noisy data, which obviously reduce the performance of the trained CNN models. Kicking out wrong labels from large-scale FR datasets is still very expensive, although some cleaning approaches are proposed. According to the analysis of the whole process of training CNN models supervised by angular margin based loss(AM-Loss) functions, we find that the $\\theta$ distribution of training samples implic"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.10357","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":""},"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-17T23:50:18Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"jaqT6U9VOU/pKmsRUj998kK84uK2G6QDImaOfJswOatd/XE8mZqNL/OQE90J/qX78OuFdQaOSIk2IXLRuuH6Dg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-29T18:22:22.759100Z"},"content_sha256":"dca93f1d233b259634fda0fbcdf86c82673d9a30392391b0d5e92b64ee4fa034","schema_version":"1.0","event_id":"sha256:dca93f1d233b259634fda0fbcdf86c82673d9a30392391b0d5e92b64ee4fa034"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/bundle.json","state_url":"https://pith.science/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-29T18:22:22Z","links":{"resolver":"https://pith.science/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5","bundle":"https://pith.science/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/bundle.json","state":"https://pith.science/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WTYXWUKBKFAEWZOEYX6WJB4BV5/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:WTYXWUKBKFAEWZOEYX6WJB4BV5","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"534d1180e9fb219ca40fe2a6493da72e4aa1dc39e2b67fef8a7518471eb42678","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-03-25T14:16:06Z","title_canon_sha256":"56a85dfedc30952587f0bce3733662a5b76cd37ec70c9cf7c2bb1b4e78bc2dbc"},"schema_version":"1.0","source":{"id":"1903.10357","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1903.10357","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"arxiv_version","alias_value":"1903.10357v2","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1903.10357","created_at":"2026-05-17T23:50:18Z"},{"alias_kind":"pith_short_12","alias_value":"WTYXWUKBKFAE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_16","alias_value":"WTYXWUKBKFAEWZOE","created_at":"2026-05-18T12:33:30Z"},{"alias_kind":"pith_short_8","alias_value":"WTYXWUKB","created_at":"2026-05-18T12:33:30Z"}],"graph_snapshots":[{"event_id":"sha256:dca93f1d233b259634fda0fbcdf86c82673d9a30392391b0d5e92b64ee4fa034","target":"graph","created_at":"2026-05-17T23:50:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"paper":{"abstract_excerpt":"Benefit from large-scale training datasets, deep Convolutional Neural Networks(CNNs) have achieved impressive results in face recognition(FR). However, tremendous scale of datasets inevitably lead to noisy data, which obviously reduce the performance of the trained CNN models. Kicking out wrong labels from large-scale FR datasets is still very expensive, although some cleaning approaches are proposed. According to the analysis of the whole process of training CNN models supervised by angular margin based loss(AM-Loss) functions, we find that the $\\theta$ distribution of training samples implic","authors_text":"Fan Zhang, Ruirui Li, Wei Hu, Yangyu Huang","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-03-25T14:16:06Z","title":"Noise-Tolerant Paradigm for Training Face Recognition CNNs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1903.10357","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0d4c9db1690c3868151672d0d913d94a5f8a3102c0a53350b8cb69ed00d6079f","target":"record","created_at":"2026-05-17T23:50:18Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"534d1180e9fb219ca40fe2a6493da72e4aa1dc39e2b67fef8a7518471eb42678","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-03-25T14:16:06Z","title_canon_sha256":"56a85dfedc30952587f0bce3733662a5b76cd37ec70c9cf7c2bb1b4e78bc2dbc"},"schema_version":"1.0","source":{"id":"1903.10357","kind":"arxiv","version":2}},"canonical_sha256":"b4f17b514151404b65c4c5fd648781af54e2f387470a8c5700c57ca1c6b5860b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b4f17b514151404b65c4c5fd648781af54e2f387470a8c5700c57ca1c6b5860b","first_computed_at":"2026-05-17T23:50:18.626642Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:50:18.626642Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"8MFq+tfvJfHMDWRT21AKpZG71R8/QyGvn53+hw6STlmZY3X1icdTZ7cqNgQZlEq6wtp5mhE+gUzv0zOl2sm+DA==","signature_status":"signed_v1","signed_at":"2026-05-17T23:50:18.627269Z","signed_message":"canonical_sha256_bytes"},"source_id":"1903.10357","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0d4c9db1690c3868151672d0d913d94a5f8a3102c0a53350b8cb69ed00d6079f","sha256:dca93f1d233b259634fda0fbcdf86c82673d9a30392391b0d5e92b64ee4fa034"],"state_sha256":"dfb18e25cf75d8388f1e3cf837953565a036de8521a30c2e6f69aa449fa1abaa"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/MxMS0iW2GE5nMID6c4EQrKK6u1/kMyPbBkUSybjo5LCKeRFkE6RSWe7BsxE9+pB8j03tY+SCdphYayrofffCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-29T18:22:22.762762Z","bundle_sha256":"9dfe7cd1104e1b487e73dffed6215d0e0d2eddea4c21cc425d0d7bbeaa0acbe4"}}