{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:G2QZWPDKDOZL5WSRCQQ35TQ36D","short_pith_number":"pith:G2QZWPDK","schema_version":"1.0","canonical_sha256":"36a19b3c6a1bb2beda511421bece1bf0f86781e325a715d65eeb2692bb2fddb2","source":{"kind":"arxiv","id":"2305.09028","version":2},"attestation_state":"computed","paper":{"title":"SKI to go Faster: Accelerating Toeplitz Neural Networks via Asymmetric Kernels","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Moreno, Jonathan Mei, Luke Walters","submitted_at":"2023-05-15T21:25:35Z","abstract_excerpt":"Toeplitz Neural Networks (TNNs) (Qin et. al. 2023) are a recent sequence model with impressive results. They require O(n log n) computational complexity and O(n) relative positional encoder (RPE) multi-layer perceptron (MLP) and decay bias calls. We aim to reduce both. We first note that the RPE is a non-SPD (symmetric positive definite) kernel and the Toeplitz matrices are pseudo-Gram matrices. Further 1) the learned kernels display spiky behavior near the main diagonals with otherwise smooth behavior; 2) the RPE MLP is slow. For bidirectional models, this motivates a sparse plus low-rank Toe"},"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":"2305.09028","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"stat.ML","submitted_at":"2023-05-15T21:25:35Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"54aff891ce5a9ba539fdc6844ad4b6cc11ea19b574ae74d403ec1f8f34c6066a","abstract_canon_sha256":"5bf4904be4fdb03f07a9397a2a11f0b4f3be42cb56a42492c469d03e0703fee6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:29:01.786479Z","signature_b64":"vrt4TKCpqvgNROgOcONLfVtKZiQ2jQdZ8KdeZUPX2NrNpitCF4ieypxWUTtejAOrGBNBE8XRiWyXvxpdoVCxDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"36a19b3c6a1bb2beda511421bece1bf0f86781e325a715d65eeb2692bb2fddb2","last_reissued_at":"2026-07-05T06:29:01.786000Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:29:01.786000Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SKI to go Faster: Accelerating Toeplitz Neural Networks via Asymmetric Kernels","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Alexander Moreno, Jonathan Mei, Luke Walters","submitted_at":"2023-05-15T21:25:35Z","abstract_excerpt":"Toeplitz Neural Networks (TNNs) (Qin et. al. 2023) are a recent sequence model with impressive results. They require O(n log n) computational complexity and O(n) relative positional encoder (RPE) multi-layer perceptron (MLP) and decay bias calls. We aim to reduce both. We first note that the RPE is a non-SPD (symmetric positive definite) kernel and the Toeplitz matrices are pseudo-Gram matrices. Further 1) the learned kernels display spiky behavior near the main diagonals with otherwise smooth behavior; 2) the RPE MLP is slow. For bidirectional models, this motivates a sparse plus low-rank Toe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.09028","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/2305.09028/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":"2305.09028","created_at":"2026-07-05T06:29:01.786057+00:00"},{"alias_kind":"arxiv_version","alias_value":"2305.09028v2","created_at":"2026-07-05T06:29:01.786057+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.09028","created_at":"2026-07-05T06:29:01.786057+00:00"},{"alias_kind":"pith_short_12","alias_value":"G2QZWPDKDOZL","created_at":"2026-07-05T06:29:01.786057+00:00"},{"alias_kind":"pith_short_16","alias_value":"G2QZWPDKDOZL5WSR","created_at":"2026-07-05T06:29:01.786057+00:00"},{"alias_kind":"pith_short_8","alias_value":"G2QZWPDK","created_at":"2026-07-05T06:29:01.786057+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.00298","citing_title":"The Price of Linear Time: Error Analysis of Structured Kernel Interpolation","ref_index":23,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D","json":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D.json","graph_json":"https://pith.science/api/pith-number/G2QZWPDKDOZL5WSRCQQ35TQ36D/graph.json","events_json":"https://pith.science/api/pith-number/G2QZWPDKDOZL5WSRCQQ35TQ36D/events.json","paper":"https://pith.science/paper/G2QZWPDK"},"agent_actions":{"view_html":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D","download_json":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D.json","view_paper":"https://pith.science/paper/G2QZWPDK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2305.09028&json=true","fetch_graph":"https://pith.science/api/pith-number/G2QZWPDKDOZL5WSRCQQ35TQ36D/graph.json","fetch_events":"https://pith.science/api/pith-number/G2QZWPDKDOZL5WSRCQQ35TQ36D/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D/action/timestamp_anchor","attest_storage":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D/action/storage_attestation","attest_author":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D/action/author_attestation","sign_citation":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D/action/citation_signature","submit_replication":"https://pith.science/pith/G2QZWPDKDOZL5WSRCQQ35TQ36D/action/replication_record"}},"created_at":"2026-07-05T06:29:01.786057+00:00","updated_at":"2026-07-05T06:29:01.786057+00:00"}