{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZAHHYWYIEDO3MZDRMD2CXXWGOY","short_pith_number":"pith:ZAHHYWYI","schema_version":"1.0","canonical_sha256":"c80e7c5b0820ddb6647160f42bdec67625a7d932f674e395b88fc163b46c117a","source":{"kind":"arxiv","id":"2602.10056","version":2},"attestation_state":"computed","paper":{"title":"WildCat: Near-Linear Attention in Theory and Practice","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Lester Mackey, Tobias Schr\\\"oder","submitted_at":"2026-02-10T18:22:32Z","abstract_excerpt":"We introduce WildCat, a high-accuracy, low-cost approach to compressing the attention mechanism in neural networks. While attention is a staple of modern network architectures, it is also notoriously expensive to deploy due to resource requirements that scale quadratically with the input sequence length $n$. WildCat avoids these quadratic costs by only attending over a small weighted coreset. Crucially, we select the coreset using a fast but spectrally-accurate subsampling algorithm -- randomly pivoted Cholesky -- and weight the elements optimally to minimise reconstruction error. Remarkably, "},"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":"2602.10056","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2026-02-10T18:22:32Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"917cade347751bb646a9b56c8c949145507b8972e4b0df2358c2b807b453102e","abstract_canon_sha256":"58820b9b1468242d1b2cbf985b6567f516fa572c483e5871974a2da04c8ed519"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T03:04:38.578266Z","signature_b64":"nU+xkJ6C4HgJx/9JBDXyiwweLVxGS3uaNTVQh2nVn3lbHZ4KCf5DbCkgVcZ/SdgJy5O+i9A9fImXpjCTeg6aBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c80e7c5b0820ddb6647160f42bdec67625a7d932f674e395b88fc163b46c117a","last_reissued_at":"2026-06-02T03:04:38.577732Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T03:04:38.577732Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"WildCat: Near-Linear Attention in Theory and Practice","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Lester Mackey, Tobias Schr\\\"oder","submitted_at":"2026-02-10T18:22:32Z","abstract_excerpt":"We introduce WildCat, a high-accuracy, low-cost approach to compressing the attention mechanism in neural networks. While attention is a staple of modern network architectures, it is also notoriously expensive to deploy due to resource requirements that scale quadratically with the input sequence length $n$. WildCat avoids these quadratic costs by only attending over a small weighted coreset. Crucially, we select the coreset using a fast but spectrally-accurate subsampling algorithm -- randomly pivoted Cholesky -- and weight the elements optimally to minimise reconstruction error. Remarkably, "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.10056","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/2602.10056/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":"2602.10056","created_at":"2026-06-02T03:04:38.577792+00:00"},{"alias_kind":"arxiv_version","alias_value":"2602.10056v2","created_at":"2026-06-02T03:04:38.577792+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.10056","created_at":"2026-06-02T03:04:38.577792+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZAHHYWYIEDO3","created_at":"2026-06-02T03:04:38.577792+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZAHHYWYIEDO3MZDR","created_at":"2026-06-02T03:04:38.577792+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZAHHYWYI","created_at":"2026-06-02T03:04:38.577792+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.05602","citing_title":"Nearly Optimal Attention Coresets","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY","json":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY.json","graph_json":"https://pith.science/api/pith-number/ZAHHYWYIEDO3MZDRMD2CXXWGOY/graph.json","events_json":"https://pith.science/api/pith-number/ZAHHYWYIEDO3MZDRMD2CXXWGOY/events.json","paper":"https://pith.science/paper/ZAHHYWYI"},"agent_actions":{"view_html":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY","download_json":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY.json","view_paper":"https://pith.science/paper/ZAHHYWYI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2602.10056&json=true","fetch_graph":"https://pith.science/api/pith-number/ZAHHYWYIEDO3MZDRMD2CXXWGOY/graph.json","fetch_events":"https://pith.science/api/pith-number/ZAHHYWYIEDO3MZDRMD2CXXWGOY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY/action/storage_attestation","attest_author":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY/action/author_attestation","sign_citation":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY/action/citation_signature","submit_replication":"https://pith.science/pith/ZAHHYWYIEDO3MZDRMD2CXXWGOY/action/replication_record"}},"created_at":"2026-06-02T03:04:38.577792+00:00","updated_at":"2026-06-02T03:04:38.577792+00:00"}