{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:FKT4BPDQJ3QVVL2XM57YJPXLRN","short_pith_number":"pith:FKT4BPDQ","schema_version":"1.0","canonical_sha256":"2aa7c0bc704ee15aaf57677f84beeb8b555c2a7c9a5e4262ed3764f37acc9f3a","source":{"kind":"arxiv","id":"2109.04553","version":2},"attestation_state":"computed","paper":{"title":"Is Attention Better Than Matrix Decomposition?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Hongxu Chen, Ke Wei, Meng-Hao Guo, Xia Li, Zhengyang Geng, Zhouchen Lin","submitted_at":"2021-09-09T20:40:19Z","abstract_excerpt":"As an essential ingredient of modern deep learning, attention mechanism, especially self-attention, plays a vital role in the global correlation discovery. However, is hand-crafted attention irreplaceable when modeling the global context? Our intriguing finding is that self-attention is not better than the matrix decomposition (MD) model developed 20 years ago regarding the performance and computational cost for encoding the long-distance dependencies. We model the global context issue as a low-rank recovery problem and show that its optimization algorithms can help design global information b"},"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":"2109.04553","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2021-09-09T20:40:19Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"1327c0e9ffe1ff79376b9f4b4322b2c7b2f8a9efc5811caa610db5a3929f450c","abstract_canon_sha256":"61d1c06231303f70f8e31f0b004ce3f7d10d77e61717da61d017c7305bebd77e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:44:12.887593Z","signature_b64":"RmbXXuBAB21RN8MNv3KZE/SXfbUiehWm+ULGr5px2/kWmzYzT0WpU0ytPjckjR1RflipE4Fg9gxq+rFKXYFWDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2aa7c0bc704ee15aaf57677f84beeb8b555c2a7c9a5e4262ed3764f37acc9f3a","last_reissued_at":"2026-07-05T03:44:12.887062Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:44:12.887062Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Is Attention Better Than Matrix Decomposition?","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Hongxu Chen, Ke Wei, Meng-Hao Guo, Xia Li, Zhengyang Geng, Zhouchen Lin","submitted_at":"2021-09-09T20:40:19Z","abstract_excerpt":"As an essential ingredient of modern deep learning, attention mechanism, especially self-attention, plays a vital role in the global correlation discovery. However, is hand-crafted attention irreplaceable when modeling the global context? Our intriguing finding is that self-attention is not better than the matrix decomposition (MD) model developed 20 years ago regarding the performance and computational cost for encoding the long-distance dependencies. We model the global context issue as a low-rank recovery problem and show that its optimization algorithms can help design global information b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2109.04553","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/2109.04553/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":"2109.04553","created_at":"2026-07-05T03:44:12.887126+00:00"},{"alias_kind":"arxiv_version","alias_value":"2109.04553v2","created_at":"2026-07-05T03:44:12.887126+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2109.04553","created_at":"2026-07-05T03:44:12.887126+00:00"},{"alias_kind":"pith_short_12","alias_value":"FKT4BPDQJ3QV","created_at":"2026-07-05T03:44:12.887126+00:00"},{"alias_kind":"pith_short_16","alias_value":"FKT4BPDQJ3QVVL2X","created_at":"2026-07-05T03:44:12.887126+00:00"},{"alias_kind":"pith_short_8","alias_value":"FKT4BPDQ","created_at":"2026-07-05T03:44:12.887126+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2607.01870","citing_title":"CamoNAS: Neural Architecture Search for Enhanced Camouflaged Object Detection","ref_index":75,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN","json":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN.json","graph_json":"https://pith.science/api/pith-number/FKT4BPDQJ3QVVL2XM57YJPXLRN/graph.json","events_json":"https://pith.science/api/pith-number/FKT4BPDQJ3QVVL2XM57YJPXLRN/events.json","paper":"https://pith.science/paper/FKT4BPDQ"},"agent_actions":{"view_html":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN","download_json":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN.json","view_paper":"https://pith.science/paper/FKT4BPDQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2109.04553&json=true","fetch_graph":"https://pith.science/api/pith-number/FKT4BPDQJ3QVVL2XM57YJPXLRN/graph.json","fetch_events":"https://pith.science/api/pith-number/FKT4BPDQJ3QVVL2XM57YJPXLRN/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN/action/storage_attestation","attest_author":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN/action/author_attestation","sign_citation":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN/action/citation_signature","submit_replication":"https://pith.science/pith/FKT4BPDQJ3QVVL2XM57YJPXLRN/action/replication_record"}},"created_at":"2026-07-05T03:44:12.887126+00:00","updated_at":"2026-07-05T03:44:12.887126+00:00"}