{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:CIC3KOK5ZYPDPL56N666PKSTSJ","short_pith_number":"pith:CIC3KOK5","schema_version":"1.0","canonical_sha256":"1205b5395dce1e37afbe6fbde7aa53927f6919ee1badf980843ff61d5b431f83","source":{"kind":"arxiv","id":"2511.20102","version":3},"attestation_state":"computed","paper":{"title":"SSA: Sparse Sparse Attention by Aligning Full and Sparse Attention Outputs in Feature Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Di Yin, Jiazheng Li, Junru Lu, Lin Gui, Xing Sun, Yulan He, Zhenyi Shen","submitted_at":"2025-11-25T09:21:57Z","abstract_excerpt":"Sparse attention reduces the quadratic complexity of full self-attention but faces two challenges: (1) an attention gap, where applying sparse attention to full-attention-trained models causes performance degradation due to train-inference distribution mismatch, and (2) a capability gap, where models trained purely with sparse attention lack complete gradient flow, preventing them from matching full-attention performance. We propose SSA (Sparse Sparse Attention), a training framework that integrates both sparse and full attention with bidirectional attention-output alignment. We prove that the"},"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":"2511.20102","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-11-25T09:21:57Z","cross_cats_sorted":[],"title_canon_sha256":"db899b5559af13d43aaf306dfcef496254f64d1a208a6c7146030a2e0884f0e4","abstract_canon_sha256":"290a15a6b6a82470e43928016ba97930829135faf72e4b6f8af2f58e8479d4c4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-04T01:09:39.866217Z","signature_b64":"TcJXVD93CrdxCbcpJasokIOa9PE09Iw0gtkUwWztNCPKUNoYNRLiZxp4/1MsQPwLJb9yMwlfvLBFTDnwlV0nAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1205b5395dce1e37afbe6fbde7aa53927f6919ee1badf980843ff61d5b431f83","last_reissued_at":"2026-06-04T01:09:39.865619Z","signature_status":"signed_v1","first_computed_at":"2026-06-04T01:09:39.865619Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SSA: Sparse Sparse Attention by Aligning Full and Sparse Attention Outputs in Feature Space","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Di Yin, Jiazheng Li, Junru Lu, Lin Gui, Xing Sun, Yulan He, Zhenyi Shen","submitted_at":"2025-11-25T09:21:57Z","abstract_excerpt":"Sparse attention reduces the quadratic complexity of full self-attention but faces two challenges: (1) an attention gap, where applying sparse attention to full-attention-trained models causes performance degradation due to train-inference distribution mismatch, and (2) a capability gap, where models trained purely with sparse attention lack complete gradient flow, preventing them from matching full-attention performance. We propose SSA (Sparse Sparse Attention), a training framework that integrates both sparse and full attention with bidirectional attention-output alignment. We prove that the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2511.20102","kind":"arxiv","version":3},"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/2511.20102/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":"2511.20102","created_at":"2026-06-04T01:09:39.865698+00:00"},{"alias_kind":"arxiv_version","alias_value":"2511.20102v3","created_at":"2026-06-04T01:09:39.865698+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2511.20102","created_at":"2026-06-04T01:09:39.865698+00:00"},{"alias_kind":"pith_short_12","alias_value":"CIC3KOK5ZYPD","created_at":"2026-06-04T01:09:39.865698+00:00"},{"alias_kind":"pith_short_16","alias_value":"CIC3KOK5ZYPDPL56","created_at":"2026-06-04T01:09:39.865698+00:00"},{"alias_kind":"pith_short_8","alias_value":"CIC3KOK5","created_at":"2026-06-04T01:09:39.865698+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ","json":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ.json","graph_json":"https://pith.science/api/pith-number/CIC3KOK5ZYPDPL56N666PKSTSJ/graph.json","events_json":"https://pith.science/api/pith-number/CIC3KOK5ZYPDPL56N666PKSTSJ/events.json","paper":"https://pith.science/paper/CIC3KOK5"},"agent_actions":{"view_html":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ","download_json":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ.json","view_paper":"https://pith.science/paper/CIC3KOK5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2511.20102&json=true","fetch_graph":"https://pith.science/api/pith-number/CIC3KOK5ZYPDPL56N666PKSTSJ/graph.json","fetch_events":"https://pith.science/api/pith-number/CIC3KOK5ZYPDPL56N666PKSTSJ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ/action/storage_attestation","attest_author":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ/action/author_attestation","sign_citation":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ/action/citation_signature","submit_replication":"https://pith.science/pith/CIC3KOK5ZYPDPL56N666PKSTSJ/action/replication_record"}},"created_at":"2026-06-04T01:09:39.865698+00:00","updated_at":"2026-06-04T01:09:39.865698+00:00"}