{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:RWNPSJXDYP7BJLSNDYDPVSRZAK","short_pith_number":"pith:RWNPSJXD","schema_version":"1.0","canonical_sha256":"8d9af926e3c3fe14ae4d1e06faca390283ecb866fa49fd722181f193177e66ce","source":{"kind":"arxiv","id":"2003.08897","version":1},"attestation_state":"computed","paper":{"title":"Normalized and Geometry-Aware Self-Attention Network for Image Captioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.MM"],"primary_cat":"cs.CV","authors_text":"Hanqing Lu, Jing Liu, Longteng Guo, Peng Yao, Shichen Lu, Xinxin Zhu","submitted_at":"2020-03-19T16:54:16Z","abstract_excerpt":"Self-attention (SA) network has shown profound value in image captioning. In this paper, we improve SA from two aspects to promote the performance of image captioning. First, we propose Normalized Self-Attention (NSA), a reparameterization of SA that brings the benefits of normalization inside SA. While normalization is previously only applied outside SA, we introduce a novel normalization method and demonstrate that it is both possible and beneficial to perform it on the hidden activations inside SA. Second, to compensate for the major limit of Transformer that it fails to model the geometry "},"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":"2003.08897","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-03-19T16:54:16Z","cross_cats_sorted":["cs.CL","cs.MM"],"title_canon_sha256":"5a7021cb1d11a75f387aaf0f9db06a332690603770f1938ffaa6f150d260a00d","abstract_canon_sha256":"b6ed8c7b5b9ded48732bbb043388e4c952afa03616c73de076421bce1cd52410"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:49:21.675213Z","signature_b64":"Iw0fkK4wLxa5UILD1qsmOvZdjoEkkSGS9/Egb44OiH4tPPA8lnqT6t1BBgLH2/T9/+DacJ8rUouD9hS5+6fbDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"8d9af926e3c3fe14ae4d1e06faca390283ecb866fa49fd722181f193177e66ce","last_reissued_at":"2026-07-05T00:49:21.674799Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:49:21.674799Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Normalized and Geometry-Aware Self-Attention Network for Image Captioning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CL","cs.MM"],"primary_cat":"cs.CV","authors_text":"Hanqing Lu, Jing Liu, Longteng Guo, Peng Yao, Shichen Lu, Xinxin Zhu","submitted_at":"2020-03-19T16:54:16Z","abstract_excerpt":"Self-attention (SA) network has shown profound value in image captioning. In this paper, we improve SA from two aspects to promote the performance of image captioning. First, we propose Normalized Self-Attention (NSA), a reparameterization of SA that brings the benefits of normalization inside SA. While normalization is previously only applied outside SA, we introduce a novel normalization method and demonstrate that it is both possible and beneficial to perform it on the hidden activations inside SA. Second, to compensate for the major limit of Transformer that it fails to model the geometry "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.08897","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/2003.08897/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":"2003.08897","created_at":"2026-07-05T00:49:21.674865+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.08897v1","created_at":"2026-07-05T00:49:21.674865+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.08897","created_at":"2026-07-05T00:49:21.674865+00:00"},{"alias_kind":"pith_short_12","alias_value":"RWNPSJXDYP7B","created_at":"2026-07-05T00:49:21.674865+00:00"},{"alias_kind":"pith_short_16","alias_value":"RWNPSJXDYP7BJLSN","created_at":"2026-07-05T00:49:21.674865+00:00"},{"alias_kind":"pith_short_8","alias_value":"RWNPSJXD","created_at":"2026-07-05T00:49:21.674865+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/RWNPSJXDYP7BJLSNDYDPVSRZAK","json":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK.json","graph_json":"https://pith.science/api/pith-number/RWNPSJXDYP7BJLSNDYDPVSRZAK/graph.json","events_json":"https://pith.science/api/pith-number/RWNPSJXDYP7BJLSNDYDPVSRZAK/events.json","paper":"https://pith.science/paper/RWNPSJXD"},"agent_actions":{"view_html":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK","download_json":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK.json","view_paper":"https://pith.science/paper/RWNPSJXD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.08897&json=true","fetch_graph":"https://pith.science/api/pith-number/RWNPSJXDYP7BJLSNDYDPVSRZAK/graph.json","fetch_events":"https://pith.science/api/pith-number/RWNPSJXDYP7BJLSNDYDPVSRZAK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK/action/storage_attestation","attest_author":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK/action/author_attestation","sign_citation":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK/action/citation_signature","submit_replication":"https://pith.science/pith/RWNPSJXDYP7BJLSNDYDPVSRZAK/action/replication_record"}},"created_at":"2026-07-05T00:49:21.674865+00:00","updated_at":"2026-07-05T00:49:21.674865+00:00"}