{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:3CXKWGFH5QEGRF2DHKAY6IVWQG","short_pith_number":"pith:3CXKWGFH","schema_version":"1.0","canonical_sha256":"d8aeab18a7ec086897433a818f22b6819d1854dca227a55ba55447c935b3c85a","source":{"kind":"arxiv","id":"2105.01386","version":2},"attestation_state":"computed","paper":{"title":"Canonical Saliency Maps: Decoding Deep Face Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"C V Jawahar, Thrupthi Ann John, Vineeth N Balasubramanian","submitted_at":"2021-05-04T09:42:56Z","abstract_excerpt":"As Deep Neural Network models for face processing tasks approach human-like performance, their deployment in critical applications such as law enforcement and access control has seen an upswing, where any failure may have far-reaching consequences. We need methods to build trust in deployed systems by making their working as transparent as possible. Existing visualization algorithms are designed for object recognition and do not give insightful results when applied to the face domain. In this work, we present 'Canonical Saliency Maps', a new method that highlights relevant facial areas by proj"},"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":"2105.01386","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-05-04T09:42:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"7aa2a736c3501a36835cd28bd02855aee62afecdfda3720bf31c07a9ae9b467a","abstract_canon_sha256":"e132cfa08f21241828e1a1acb149d0271cf41c4bfcc0ef35ecf668a8a1977a7d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:05:49.882634Z","signature_b64":"PAGXbbJWwnFPyiEk2ndDsoCTQUarFJAeCK7pWr7LX/b8lnypXmFxvp0a8Az/lWHvXFR1zYD/xDwkDXAykZM9Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d8aeab18a7ec086897433a818f22b6819d1854dca227a55ba55447c935b3c85a","last_reissued_at":"2026-07-05T03:05:49.882171Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:05:49.882171Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Canonical Saliency Maps: Decoding Deep Face Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"C V Jawahar, Thrupthi Ann John, Vineeth N Balasubramanian","submitted_at":"2021-05-04T09:42:56Z","abstract_excerpt":"As Deep Neural Network models for face processing tasks approach human-like performance, their deployment in critical applications such as law enforcement and access control has seen an upswing, where any failure may have far-reaching consequences. We need methods to build trust in deployed systems by making their working as transparent as possible. Existing visualization algorithms are designed for object recognition and do not give insightful results when applied to the face domain. In this work, we present 'Canonical Saliency Maps', a new method that highlights relevant facial areas by proj"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.01386","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/2105.01386/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":"2105.01386","created_at":"2026-07-05T03:05:49.882232+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.01386v2","created_at":"2026-07-05T03:05:49.882232+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.01386","created_at":"2026-07-05T03:05:49.882232+00:00"},{"alias_kind":"pith_short_12","alias_value":"3CXKWGFH5QEG","created_at":"2026-07-05T03:05:49.882232+00:00"},{"alias_kind":"pith_short_16","alias_value":"3CXKWGFH5QEGRF2D","created_at":"2026-07-05T03:05:49.882232+00:00"},{"alias_kind":"pith_short_8","alias_value":"3CXKWGFH","created_at":"2026-07-05T03:05:49.882232+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/3CXKWGFH5QEGRF2DHKAY6IVWQG","json":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG.json","graph_json":"https://pith.science/api/pith-number/3CXKWGFH5QEGRF2DHKAY6IVWQG/graph.json","events_json":"https://pith.science/api/pith-number/3CXKWGFH5QEGRF2DHKAY6IVWQG/events.json","paper":"https://pith.science/paper/3CXKWGFH"},"agent_actions":{"view_html":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG","download_json":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG.json","view_paper":"https://pith.science/paper/3CXKWGFH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.01386&json=true","fetch_graph":"https://pith.science/api/pith-number/3CXKWGFH5QEGRF2DHKAY6IVWQG/graph.json","fetch_events":"https://pith.science/api/pith-number/3CXKWGFH5QEGRF2DHKAY6IVWQG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG/action/storage_attestation","attest_author":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG/action/author_attestation","sign_citation":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG/action/citation_signature","submit_replication":"https://pith.science/pith/3CXKWGFH5QEGRF2DHKAY6IVWQG/action/replication_record"}},"created_at":"2026-07-05T03:05:49.882232+00:00","updated_at":"2026-07-05T03:05:49.882232+00:00"}