{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:2NIY22DPWQ2M2GMRX2Y4NBAW76","short_pith_number":"pith:2NIY22DP","schema_version":"1.0","canonical_sha256":"d3518d686fb434cd1991beb1c68416ff8a629ae72b75956c78d6942188d52dac","source":{"kind":"arxiv","id":"2405.17381","version":2},"attestation_state":"computed","paper":{"title":"Various Lengths, Constant Speed: Efficient Language Modeling with Lightning Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dong Li, Weigao Sun, Weixuan Sun, Xuyang Shen, Yiran Zhong, Zhen Qin","submitted_at":"2024-05-27T17:38:13Z","abstract_excerpt":"We present Lightning Attention, the first linear attention implementation that maintains a constant training speed for various sequence lengths under fixed memory consumption. Due to the issue with cumulative summation operations (cumsum), previous linear attention implementations cannot achieve their theoretical advantage in a casual setting. However, this issue can be effectively solved by utilizing different attention calculation strategies to compute the different parts of attention. Specifically, we split the attention calculation into intra-blocks and inter-blocks and use conventional at"},"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":"2405.17381","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2024-05-27T17:38:13Z","cross_cats_sorted":[],"title_canon_sha256":"91a98cc495940a020d6fe5d64ba38d0891fb3b777af9fbe8cfe5f541d478ee5c","abstract_canon_sha256":"7c780eac930e3d60bfa9f802b2cdf114a78ee303e705f984f7fea3955bd8e077"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:34:19.531498Z","signature_b64":"m4bM3hkg1xG0+FhSlHRvnGxkE8N2XHNIm4ob6i3MPYGx4ZRqxYy1AI2+ZTmlVRmVbaCMZHqbAnhe1yhWT3ZTDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d3518d686fb434cd1991beb1c68416ff8a629ae72b75956c78d6942188d52dac","last_reissued_at":"2026-07-05T08:34:19.531067Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:34:19.531067Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Various Lengths, Constant Speed: Efficient Language Modeling with Lightning Attention","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Dong Li, Weigao Sun, Weixuan Sun, Xuyang Shen, Yiran Zhong, Zhen Qin","submitted_at":"2024-05-27T17:38:13Z","abstract_excerpt":"We present Lightning Attention, the first linear attention implementation that maintains a constant training speed for various sequence lengths under fixed memory consumption. Due to the issue with cumulative summation operations (cumsum), previous linear attention implementations cannot achieve their theoretical advantage in a casual setting. However, this issue can be effectively solved by utilizing different attention calculation strategies to compute the different parts of attention. Specifically, we split the attention calculation into intra-blocks and inter-blocks and use conventional at"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.17381","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/2405.17381/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":"2405.17381","created_at":"2026-07-05T08:34:19.531137+00:00"},{"alias_kind":"arxiv_version","alias_value":"2405.17381v2","created_at":"2026-07-05T08:34:19.531137+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.17381","created_at":"2026-07-05T08:34:19.531137+00:00"},{"alias_kind":"pith_short_12","alias_value":"2NIY22DPWQ2M","created_at":"2026-07-05T08:34:19.531137+00:00"},{"alias_kind":"pith_short_16","alias_value":"2NIY22DPWQ2M2GMR","created_at":"2026-07-05T08:34:19.531137+00:00"},{"alias_kind":"pith_short_8","alias_value":"2NIY22DP","created_at":"2026-07-05T08:34:19.531137+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08587","citing_title":"Kaczmarz Linear Attention","ref_index":26,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76","json":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76.json","graph_json":"https://pith.science/api/pith-number/2NIY22DPWQ2M2GMRX2Y4NBAW76/graph.json","events_json":"https://pith.science/api/pith-number/2NIY22DPWQ2M2GMRX2Y4NBAW76/events.json","paper":"https://pith.science/paper/2NIY22DP"},"agent_actions":{"view_html":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76","download_json":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76.json","view_paper":"https://pith.science/paper/2NIY22DP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2405.17381&json=true","fetch_graph":"https://pith.science/api/pith-number/2NIY22DPWQ2M2GMRX2Y4NBAW76/graph.json","fetch_events":"https://pith.science/api/pith-number/2NIY22DPWQ2M2GMRX2Y4NBAW76/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76/action/storage_attestation","attest_author":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76/action/author_attestation","sign_citation":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76/action/citation_signature","submit_replication":"https://pith.science/pith/2NIY22DPWQ2M2GMRX2Y4NBAW76/action/replication_record"}},"created_at":"2026-07-05T08:34:19.531137+00:00","updated_at":"2026-07-05T08:34:19.531137+00:00"}