{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2020:M675XSCI33DWVVJIQJJOD3R6YR","short_pith_number":"pith:M675XSCI","schema_version":"1.0","canonical_sha256":"67bfdbc848dec76ad5288252e1ee3ec4530f6043d39cd2a3a37765a9ba382548","source":{"kind":"arxiv","id":"2003.07853","version":2},"attestation_state":"computed","paper":{"title":"Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Bradley Green, Hartwig Adam, Huiyu Wang, Liang-Chieh Chen, Yukun Zhu","submitted_at":"2020-03-17T17:59:56Z","abstract_excerpt":"Convolution exploits locality for efficiency at a cost of missing long range context. Self-attention has been adopted to augment CNNs with non-local interactions. Recent works prove it possible to stack self-attention layers to obtain a fully attentional network by restricting the attention to a local region. In this paper, we attempt to remove this constraint by factorizing 2D self-attention into two 1D self-attentions. This reduces computation complexity and allows performing attention within a larger or even global region. In companion, we also propose a position-sensitive self-attention de"},"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":"2003.07853","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2020-03-17T17:59:56Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1f0944ede22bc986918b07cbca96a792dac2980f3c7ab5e424c67fd56ccc0d35","abstract_canon_sha256":"f816118614f309e0d1bccd9d0ac20d0ce17239ad19798fa4bc78d19a85a7ccf0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:25:25.109577Z","signature_b64":"3WO8N0r+pNS5r1hvKL4I7pIVYGVtkRngDbtlAm/fILn/00mLQQPVb1LRC5pDMMoP+yCyy2abcn9Q8IuR8+pJDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"67bfdbc848dec76ad5288252e1ee3ec4530f6043d39cd2a3a37765a9ba382548","last_reissued_at":"2026-07-05T01:25:25.109072Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:25:25.109072Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Axial-DeepLab: Stand-Alone Axial-Attention for Panoptic Segmentation","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Alan Yuille, Bradley Green, Hartwig Adam, Huiyu Wang, Liang-Chieh Chen, Yukun Zhu","submitted_at":"2020-03-17T17:59:56Z","abstract_excerpt":"Convolution exploits locality for efficiency at a cost of missing long range context. Self-attention has been adopted to augment CNNs with non-local interactions. Recent works prove it possible to stack self-attention layers to obtain a fully attentional network by restricting the attention to a local region. In this paper, we attempt to remove this constraint by factorizing 2D self-attention into two 1D self-attentions. This reduces computation complexity and allows performing attention within a larger or even global region. In companion, we also propose a position-sensitive self-attention de"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2003.07853","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/2003.07853/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":"2003.07853","created_at":"2026-07-05T01:25:25.109132+00:00"},{"alias_kind":"arxiv_version","alias_value":"2003.07853v2","created_at":"2026-07-05T01:25:25.109132+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2003.07853","created_at":"2026-07-05T01:25:25.109132+00:00"},{"alias_kind":"pith_short_12","alias_value":"M675XSCI33DW","created_at":"2026-07-05T01:25:25.109132+00:00"},{"alias_kind":"pith_short_16","alias_value":"M675XSCI33DWVVJI","created_at":"2026-07-05T01:25:25.109132+00:00"},{"alias_kind":"pith_short_8","alias_value":"M675XSCI","created_at":"2026-07-05T01:25:25.109132+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2010.11929","citing_title":"An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2010.04159","citing_title":"Deformable DETR: Deformable Transformers for End-to-End Object Detection","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR","json":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR.json","graph_json":"https://pith.science/api/pith-number/M675XSCI33DWVVJIQJJOD3R6YR/graph.json","events_json":"https://pith.science/api/pith-number/M675XSCI33DWVVJIQJJOD3R6YR/events.json","paper":"https://pith.science/paper/M675XSCI"},"agent_actions":{"view_html":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR","download_json":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR.json","view_paper":"https://pith.science/paper/M675XSCI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2003.07853&json=true","fetch_graph":"https://pith.science/api/pith-number/M675XSCI33DWVVJIQJJOD3R6YR/graph.json","fetch_events":"https://pith.science/api/pith-number/M675XSCI33DWVVJIQJJOD3R6YR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR/action/storage_attestation","attest_author":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR/action/author_attestation","sign_citation":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR/action/citation_signature","submit_replication":"https://pith.science/pith/M675XSCI33DWVVJIQJJOD3R6YR/action/replication_record"}},"created_at":"2026-07-05T01:25:25.109132+00:00","updated_at":"2026-07-05T01:25:25.109132+00:00"}