{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:BQDSYDBBK4O5A5VPW6RA6UDI3D","short_pith_number":"pith:BQDSYDBB","schema_version":"1.0","canonical_sha256":"0c072c0c21571dd076afb7a20f5068d8dfdd2db723929f0a76c546d8197768ae","source":{"kind":"arxiv","id":"2406.16747","version":1},"attestation_state":"computed","paper":{"title":"Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Lou, Kewei Tu, Zilong Zheng, Zixia Jia","submitted_at":"2024-06-24T15:55:59Z","abstract_excerpt":"Accommodating long sequences efficiently in autoregressive Transformers, especially within an extended context window, poses significant challenges due to the quadratic computational complexity and substantial KV memory requirements inherent in self-attention mechanisms. In this work, we introduce SPARSEK Attention, a novel sparse attention mechanism designed to overcome these computational and memory obstacles while maintaining performance. Our approach integrates a scoring network and a differentiable top-k mask operator, SPARSEK, to select a constant number of KV pairs for each query, there"},"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":"2406.16747","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-06-24T15:55:59Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"3b732979ff46c062500ae0e99a0954353df3cec9dfe377f8f6d8301f01d55c8d","abstract_canon_sha256":"19911dcd8433f97316fb98f1965026cbe2c12198ebb62a4c99909096d02a5fd3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:36:03.147272Z","signature_b64":"uEcQ3dmegf8tMFW8Z7wn2bDJRVlXX0i651V3x/eFR8RZAZlzvBpRR5JkBIxh9IZnGbtjXhCRmWrWxeDs41jLCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c072c0c21571dd076afb7a20f5068d8dfdd2db723929f0a76c546d8197768ae","last_reissued_at":"2026-07-05T08:36:03.146838Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:36:03.146838Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Sparser is Faster and Less is More: Efficient Sparse Attention for Long-Range Transformers","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Chao Lou, Kewei Tu, Zilong Zheng, Zixia Jia","submitted_at":"2024-06-24T15:55:59Z","abstract_excerpt":"Accommodating long sequences efficiently in autoregressive Transformers, especially within an extended context window, poses significant challenges due to the quadratic computational complexity and substantial KV memory requirements inherent in self-attention mechanisms. In this work, we introduce SPARSEK Attention, a novel sparse attention mechanism designed to overcome these computational and memory obstacles while maintaining performance. Our approach integrates a scoring network and a differentiable top-k mask operator, SPARSEK, to select a constant number of KV pairs for each query, there"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2406.16747","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/2406.16747/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":"2406.16747","created_at":"2026-07-05T08:36:03.146896+00:00"},{"alias_kind":"arxiv_version","alias_value":"2406.16747v1","created_at":"2026-07-05T08:36:03.146896+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2406.16747","created_at":"2026-07-05T08:36:03.146896+00:00"},{"alias_kind":"pith_short_12","alias_value":"BQDSYDBBK4O5","created_at":"2026-07-05T08:36:03.146896+00:00"},{"alias_kind":"pith_short_16","alias_value":"BQDSYDBBK4O5A5VP","created_at":"2026-07-05T08:36:03.146896+00:00"},{"alias_kind":"pith_short_8","alias_value":"BQDSYDBB","created_at":"2026-07-05T08:36:03.146896+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":6,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2409.06679","citing_title":"E2LLM: Encoder Elongated Large Language Models for Long-Context Understanding and Reasoning","ref_index":21,"is_internal_anchor":false},{"citing_arxiv_id":"2410.04960","citing_title":"On Efficient Variants of Segment Anything Model: A Survey","ref_index":198,"is_internal_anchor":false},{"citing_arxiv_id":"2504.02181","citing_title":"A Survey of Scaling in Large Language Model Reasoning","ref_index":122,"is_internal_anchor":false},{"citing_arxiv_id":"2605.15684","citing_title":"ElasticDiT: Efficient Diffusion Transformers via Elastic Architecture and Sparse Attention for High-Resolution Image Generation on Mobile Devices","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12851","citing_title":"Chimera: Neuro-Symbolic Attention Primitives for Trustworthy Dataplane Intelligence","ref_index":31,"is_internal_anchor":false},{"citing_arxiv_id":"2602.18109","citing_title":"TempoNet: Slack-Quantized Transformer-Guided Reinforcement Scheduler for Adaptive Deadline-Centric Real-Time Dispatchs","ref_index":19,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D","json":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D.json","graph_json":"https://pith.science/api/pith-number/BQDSYDBBK4O5A5VPW6RA6UDI3D/graph.json","events_json":"https://pith.science/api/pith-number/BQDSYDBBK4O5A5VPW6RA6UDI3D/events.json","paper":"https://pith.science/paper/BQDSYDBB"},"agent_actions":{"view_html":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D","download_json":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D.json","view_paper":"https://pith.science/paper/BQDSYDBB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2406.16747&json=true","fetch_graph":"https://pith.science/api/pith-number/BQDSYDBBK4O5A5VPW6RA6UDI3D/graph.json","fetch_events":"https://pith.science/api/pith-number/BQDSYDBBK4O5A5VPW6RA6UDI3D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D/action/storage_attestation","attest_author":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D/action/author_attestation","sign_citation":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D/action/citation_signature","submit_replication":"https://pith.science/pith/BQDSYDBBK4O5A5VPW6RA6UDI3D/action/replication_record"}},"created_at":"2026-07-05T08:36:03.146896+00:00","updated_at":"2026-07-05T08:36:03.146896+00:00"}