{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:VK7TE46SVF2HWHK446NRHAS7B3","short_pith_number":"pith:VK7TE46S","schema_version":"1.0","canonical_sha256":"aabf3273d2a9747b1d5ce79b13825f0ee58e8002275b39c476e327bd47ed3584","source":{"kind":"arxiv","id":"2311.13541","version":4},"attestation_state":"computed","paper":{"title":"Linear Log-Normal Attention with Unbiased Concentration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Emir Haleva, Joseph Kampeas, Yury Nahshan","submitted_at":"2023-11-22T17:30:41Z","abstract_excerpt":"Transformer models have achieved remarkable results in a wide range of applications. However, their scalability is hampered by the quadratic time and memory complexity of the self-attention mechanism concerning the sequence length. This limitation poses a substantial obstacle when dealing with long documents or high-resolution images. In this work, we study the self-attention mechanism by analyzing the distribution of the attention matrix and its concentration ability. Furthermore, we propose instruments to measure these quantities and introduce a novel self-attention mechanism, Linear Log-Nor"},"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":"2311.13541","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-11-22T17:30:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"57b5e59cc6d62b4b3c85142a110f191723d9e751d68e75005f5f56f02fd0de7f","abstract_canon_sha256":"4a4434dff8a800f418b0eefb7d47cd517aaef9b2288d175ac912af47698d657b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:49:16.444223Z","signature_b64":"aCJN2bizzUyMYNCffyDaesyICkeox/A3+A7ps69Gt424AkCzS38wzdiiuIBe2mg+ofpSF1fNTG4mpL3LJNRdAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aabf3273d2a9747b1d5ce79b13825f0ee58e8002275b39c476e327bd47ed3584","last_reissued_at":"2026-07-05T07:49:16.443762Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:49:16.443762Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Linear Log-Normal Attention with Unbiased Concentration","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Emir Haleva, Joseph Kampeas, Yury Nahshan","submitted_at":"2023-11-22T17:30:41Z","abstract_excerpt":"Transformer models have achieved remarkable results in a wide range of applications. However, their scalability is hampered by the quadratic time and memory complexity of the self-attention mechanism concerning the sequence length. This limitation poses a substantial obstacle when dealing with long documents or high-resolution images. In this work, we study the self-attention mechanism by analyzing the distribution of the attention matrix and its concentration ability. Furthermore, we propose instruments to measure these quantities and introduce a novel self-attention mechanism, Linear Log-Nor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.13541","kind":"arxiv","version":4},"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/2311.13541/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":"2311.13541","created_at":"2026-07-05T07:49:16.443817+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.13541v4","created_at":"2026-07-05T07:49:16.443817+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.13541","created_at":"2026-07-05T07:49:16.443817+00:00"},{"alias_kind":"pith_short_12","alias_value":"VK7TE46SVF2H","created_at":"2026-07-05T07:49:16.443817+00:00"},{"alias_kind":"pith_short_16","alias_value":"VK7TE46SVF2HWHK4","created_at":"2026-07-05T07:49:16.443817+00:00"},{"alias_kind":"pith_short_8","alias_value":"VK7TE46S","created_at":"2026-07-05T07:49:16.443817+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2604.19028","citing_title":"Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priors","ref_index":275,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3","json":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3.json","graph_json":"https://pith.science/api/pith-number/VK7TE46SVF2HWHK446NRHAS7B3/graph.json","events_json":"https://pith.science/api/pith-number/VK7TE46SVF2HWHK446NRHAS7B3/events.json","paper":"https://pith.science/paper/VK7TE46S"},"agent_actions":{"view_html":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3","download_json":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3.json","view_paper":"https://pith.science/paper/VK7TE46S","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.13541&json=true","fetch_graph":"https://pith.science/api/pith-number/VK7TE46SVF2HWHK446NRHAS7B3/graph.json","fetch_events":"https://pith.science/api/pith-number/VK7TE46SVF2HWHK446NRHAS7B3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3/action/storage_attestation","attest_author":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3/action/author_attestation","sign_citation":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3/action/citation_signature","submit_replication":"https://pith.science/pith/VK7TE46SVF2HWHK446NRHAS7B3/action/replication_record"}},"created_at":"2026-07-05T07:49:16.443817+00:00","updated_at":"2026-07-05T07:49:16.443817+00:00"}