{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XNOIKLEBYQ5BPMEZGUQTUW3HVG","short_pith_number":"pith:XNOIKLEB","schema_version":"1.0","canonical_sha256":"bb5c852c81c43a17b09935213a5b67a9978425de23d6d27631a93ac5299f55c5","source":{"kind":"arxiv","id":"2505.23520","version":1},"attestation_state":"computed","paper":{"title":"AnchorAttention: Difference-Aware Sparse Attention with Stripe Granularity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dian Ding, Dong Guo, Fang Wu, Guoliang Zhu, Yiming Zhang, Yu Zhang","submitted_at":"2025-05-29T14:59:06Z","abstract_excerpt":"Large Language Models (LLMs) with extended context lengths face significant computational challenges during the pre-filling phase, primarily due to the quadratic complexity of self-attention. Existing methods typically employ dynamic pattern matching and block-sparse low-level implementations. However, their reliance on local information for pattern identification fails to capture global contexts, and the coarse granularity of blocks leads to persistent internal sparsity, resulting in suboptimal accuracy and efficiency. To address these limitations, we propose \\textbf{AnchorAttention}, a diffe"},"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":"2505.23520","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-29T14:59:06Z","cross_cats_sorted":[],"title_canon_sha256":"22a498988d8c6f9d9d1d650e330ccf8b3aa134d2008c409055f3ecfadb73ff8f","abstract_canon_sha256":"d7ae2b4706afe7838997e0961a61f4d01f8a36e970b24c7ffffb31f116f1cd02"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:12:03.368690Z","signature_b64":"PZtUB72+WzEEF70ibONCd1vW0Qtm41uWK1HmabAyukD/JOpFLZN6NlowdjpnwYgFma9RyXj19PhN4hbbT1vQBQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bb5c852c81c43a17b09935213a5b67a9978425de23d6d27631a93ac5299f55c5","last_reissued_at":"2026-07-05T11:12:03.367615Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:12:03.367615Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"AnchorAttention: Difference-Aware Sparse Attention with Stripe Granularity","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Dian Ding, Dong Guo, Fang Wu, Guoliang Zhu, Yiming Zhang, Yu Zhang","submitted_at":"2025-05-29T14:59:06Z","abstract_excerpt":"Large Language Models (LLMs) with extended context lengths face significant computational challenges during the pre-filling phase, primarily due to the quadratic complexity of self-attention. Existing methods typically employ dynamic pattern matching and block-sparse low-level implementations. However, their reliance on local information for pattern identification fails to capture global contexts, and the coarse granularity of blocks leads to persistent internal sparsity, resulting in suboptimal accuracy and efficiency. To address these limitations, we propose \\textbf{AnchorAttention}, a diffe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.23520","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/2505.23520/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":"2505.23520","created_at":"2026-07-05T11:12:03.367689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.23520v1","created_at":"2026-07-05T11:12:03.367689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.23520","created_at":"2026-07-05T11:12:03.367689+00:00"},{"alias_kind":"pith_short_12","alias_value":"XNOIKLEBYQ5B","created_at":"2026-07-05T11:12:03.367689+00:00"},{"alias_kind":"pith_short_16","alias_value":"XNOIKLEBYQ5BPMEZ","created_at":"2026-07-05T11:12:03.367689+00:00"},{"alias_kind":"pith_short_8","alias_value":"XNOIKLEB","created_at":"2026-07-05T11:12:03.367689+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.11498","citing_title":"Lag-Relative Sparse Attention In Long Context Training","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG","json":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG.json","graph_json":"https://pith.science/api/pith-number/XNOIKLEBYQ5BPMEZGUQTUW3HVG/graph.json","events_json":"https://pith.science/api/pith-number/XNOIKLEBYQ5BPMEZGUQTUW3HVG/events.json","paper":"https://pith.science/paper/XNOIKLEB"},"agent_actions":{"view_html":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG","download_json":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG.json","view_paper":"https://pith.science/paper/XNOIKLEB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.23520&json=true","fetch_graph":"https://pith.science/api/pith-number/XNOIKLEBYQ5BPMEZGUQTUW3HVG/graph.json","fetch_events":"https://pith.science/api/pith-number/XNOIKLEBYQ5BPMEZGUQTUW3HVG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG/action/storage_attestation","attest_author":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG/action/author_attestation","sign_citation":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG/action/citation_signature","submit_replication":"https://pith.science/pith/XNOIKLEBYQ5BPMEZGUQTUW3HVG/action/replication_record"}},"created_at":"2026-07-05T11:12:03.367689+00:00","updated_at":"2026-07-05T11:12:03.367689+00:00"}