{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A6ZVQGDG3FNZKWO3WXHBF3P546","short_pith_number":"pith:A6ZVQGDG","schema_version":"1.0","canonical_sha256":"07b3581866d95b9559dbb5ce12edfde78b78138ed5ef226e90d4e927994116f3","source":{"kind":"arxiv","id":"2502.07563","version":1},"attestation_state":"computed","paper":{"title":"LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Disen Lan, Weigao Sun, Xiaoye Qu, Yiran Zhong, Yu Cheng","submitted_at":"2025-02-11T14:01:39Z","abstract_excerpt":"Linear sequence modeling approaches, such as linear attention, provide advantages like linear-time training and constant-memory inference over sequence lengths. However, existing sequence parallelism (SP) methods are either not optimized for the right-product-first feature of linear attention or use a ring-style communication strategy, which results in lower computation parallelism, limits their scalability for longer sequences in distributed systems. In this paper, we introduce LASP-2, a new SP method to enhance both communication and computation parallelism when training linear attention tra"},"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":"2502.07563","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-02-11T14:01:39Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"3a34f5ffc31158a680feea7e99a11c86dbd4dc777c2766ebef0d6792ec590171","abstract_canon_sha256":"e00654b5328a8d23f7b44c0e65f4cc80c8f3bfb83f56b19671e1179b74633482"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:12:39.330290Z","signature_b64":"KaGXBHOaRBHWHg6EFc4PZLRJ8CLdB7obekwVKbbFEJfP/wBFUnvwRS/f0Icuq/2HCiYipCVZhl3b6LizpTqhCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"07b3581866d95b9559dbb5ce12edfde78b78138ed5ef226e90d4e927994116f3","last_reissued_at":"2026-07-05T10:12:39.329744Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:12:39.329744Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"LASP-2: Rethinking Sequence Parallelism for Linear Attention and Its Hybrid","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Disen Lan, Weigao Sun, Xiaoye Qu, Yiran Zhong, Yu Cheng","submitted_at":"2025-02-11T14:01:39Z","abstract_excerpt":"Linear sequence modeling approaches, such as linear attention, provide advantages like linear-time training and constant-memory inference over sequence lengths. However, existing sequence parallelism (SP) methods are either not optimized for the right-product-first feature of linear attention or use a ring-style communication strategy, which results in lower computation parallelism, limits their scalability for longer sequences in distributed systems. In this paper, we introduce LASP-2, a new SP method to enhance both communication and computation parallelism when training linear attention tra"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2502.07563","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/2502.07563/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":"2502.07563","created_at":"2026-07-05T10:12:39.329798+00:00"},{"alias_kind":"arxiv_version","alias_value":"2502.07563v1","created_at":"2026-07-05T10:12:39.329798+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2502.07563","created_at":"2026-07-05T10:12:39.329798+00:00"},{"alias_kind":"pith_short_12","alias_value":"A6ZVQGDG3FNZ","created_at":"2026-07-05T10:12:39.329798+00:00"},{"alias_kind":"pith_short_16","alias_value":"A6ZVQGDG3FNZKWO3","created_at":"2026-07-05T10:12:39.329798+00:00"},{"alias_kind":"pith_short_8","alias_value":"A6ZVQGDG","created_at":"2026-07-05T10:12:39.329798+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.05276","citing_title":"SpikingBrain: Spiking Brain-inspired Large Models","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546","json":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546.json","graph_json":"https://pith.science/api/pith-number/A6ZVQGDG3FNZKWO3WXHBF3P546/graph.json","events_json":"https://pith.science/api/pith-number/A6ZVQGDG3FNZKWO3WXHBF3P546/events.json","paper":"https://pith.science/paper/A6ZVQGDG"},"agent_actions":{"view_html":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546","download_json":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546.json","view_paper":"https://pith.science/paper/A6ZVQGDG","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2502.07563&json=true","fetch_graph":"https://pith.science/api/pith-number/A6ZVQGDG3FNZKWO3WXHBF3P546/graph.json","fetch_events":"https://pith.science/api/pith-number/A6ZVQGDG3FNZKWO3WXHBF3P546/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546/action/storage_attestation","attest_author":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546/action/author_attestation","sign_citation":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546/action/citation_signature","submit_replication":"https://pith.science/pith/A6ZVQGDG3FNZKWO3WXHBF3P546/action/replication_record"}},"created_at":"2026-07-05T10:12:39.329798+00:00","updated_at":"2026-07-05T10:12:39.329798+00:00"}