{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:LTACNFA4SCGYK3FDEK6JVMAQQO","short_pith_number":"pith:LTACNFA4","schema_version":"1.0","canonical_sha256":"5cc026941c908d856ca322bc9ab01083aa73e17d3e8c96637101bc6ee81a8932","source":{"kind":"arxiv","id":"2301.04970","version":1},"attestation_state":"computed","paper":{"title":"Hierarchical Dynamic Masks for Visual Explanation of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lianghua He, Longzhen Yang, Yihang Liu, Yitao Peng","submitted_at":"2023-01-12T12:24:49Z","abstract_excerpt":"Saliency methods generating visual explanatory maps representing the importance of image pixels for model classification is a popular technique for explaining neural network decisions. Hierarchical dynamic masks (HDM), a novel explanatory maps generation method, is proposed in this paper to enhance the granularity and comprehensiveness of saliency maps. First, we suggest the dynamic masks (DM), which enables multiple small-sized benchmark mask vectors to roughly learn the critical information in the image through an optimization method. Then the benchmark mask vectors guide the learning of lar"},"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":"2301.04970","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2023-01-12T12:24:49Z","cross_cats_sorted":[],"title_canon_sha256":"f7ef2aab0ceba103f299a7a1e4ceda849b891d0dd58c17735489d99a05a0d97e","abstract_canon_sha256":"6a38021816c38297aeeaf893ec6fb617570f6850150a08d801edcf0def4064a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:32:38.072902Z","signature_b64":"auJgojXnfPpYALCeQ623wXLA+mrcqiRC+GjNHjU1mlOqC7WwlbT821GuIUomKe6+yr/weDv54kqcwTpoLKqwBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cc026941c908d856ca322bc9ab01083aa73e17d3e8c96637101bc6ee81a8932","last_reissued_at":"2026-07-05T05:32:38.072556Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:32:38.072556Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Hierarchical Dynamic Masks for Visual Explanation of Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lianghua He, Longzhen Yang, Yihang Liu, Yitao Peng","submitted_at":"2023-01-12T12:24:49Z","abstract_excerpt":"Saliency methods generating visual explanatory maps representing the importance of image pixels for model classification is a popular technique for explaining neural network decisions. Hierarchical dynamic masks (HDM), a novel explanatory maps generation method, is proposed in this paper to enhance the granularity and comprehensiveness of saliency maps. First, we suggest the dynamic masks (DM), which enables multiple small-sized benchmark mask vectors to roughly learn the critical information in the image through an optimization method. Then the benchmark mask vectors guide the learning of lar"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.04970","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/2301.04970/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":"2301.04970","created_at":"2026-07-05T05:32:38.072612+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.04970v1","created_at":"2026-07-05T05:32:38.072612+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.04970","created_at":"2026-07-05T05:32:38.072612+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTACNFA4SCGY","created_at":"2026-07-05T05:32:38.072612+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTACNFA4SCGYK3FD","created_at":"2026-07-05T05:32:38.072612+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTACNFA4","created_at":"2026-07-05T05:32:38.072612+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/LTACNFA4SCGYK3FDEK6JVMAQQO","json":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO.json","graph_json":"https://pith.science/api/pith-number/LTACNFA4SCGYK3FDEK6JVMAQQO/graph.json","events_json":"https://pith.science/api/pith-number/LTACNFA4SCGYK3FDEK6JVMAQQO/events.json","paper":"https://pith.science/paper/LTACNFA4"},"agent_actions":{"view_html":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO","download_json":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO.json","view_paper":"https://pith.science/paper/LTACNFA4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.04970&json=true","fetch_graph":"https://pith.science/api/pith-number/LTACNFA4SCGYK3FDEK6JVMAQQO/graph.json","fetch_events":"https://pith.science/api/pith-number/LTACNFA4SCGYK3FDEK6JVMAQQO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO/action/storage_attestation","attest_author":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO/action/author_attestation","sign_citation":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO/action/citation_signature","submit_replication":"https://pith.science/pith/LTACNFA4SCGYK3FDEK6JVMAQQO/action/replication_record"}},"created_at":"2026-07-05T05:32:38.072612+00:00","updated_at":"2026-07-05T05:32:38.072612+00:00"}