{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:RFH2SRPSE5KBZTTFMYXYBBI3CZ","short_pith_number":"pith:RFH2SRPS","schema_version":"1.0","canonical_sha256":"894fa945f227541cce65662f80851b1650784250e3e9e73a8aeb8e2e7f98c77b","source":{"kind":"arxiv","id":"2410.02117","version":2},"attestation_state":"computed","paper":{"title":"Searching for Efficient Linear Layers over a Continuous Space of Structured Matrices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andres Potapczynski, Andrew Gordon Wilson, Bayan Bruss, Christopher De Sa, Christopher Ferri, Marc Finzi, Micah Goldblum, Shikai Qiu, Zixi Chen","submitted_at":"2024-10-03T00:44:50Z","abstract_excerpt":"Dense linear layers are the dominant computational bottleneck in large neural networks, presenting a critical need for more efficient alternatives. Previous efforts focused on a small number of hand-crafted structured matrices and neglected to investigate whether these structures can surpass dense layers in terms of compute-optimal scaling laws when both the model size and training examples are optimally allocated. In this work, we present a unifying framework that enables searching among all linear operators expressible via an Einstein summation. This framework encompasses many previously pro"},"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":"2410.02117","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2024-10-03T00:44:50Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"4e0b0f354237abb80bc53bff4ef98fab5c727ce3d06ff6c32c4dc82baefe8536","abstract_canon_sha256":"969f492895763308905833778d982cb398f53565ce5c876da7ecabb0109403fc"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:15:59.834757Z","signature_b64":"IvW5F4TlY1oUtpeGjIkx330rcnmBSQ2HVY1zpnriVqXovQJpdUPkYXYwISWPzNzkeUjV5D+Hx1z/9SwrId66Bw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"894fa945f227541cce65662f80851b1650784250e3e9e73a8aeb8e2e7f98c77b","last_reissued_at":"2026-07-05T09:15:59.834262Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:15:59.834262Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Searching for Efficient Linear Layers over a Continuous Space of Structured Matrices","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Andres Potapczynski, Andrew Gordon Wilson, Bayan Bruss, Christopher De Sa, Christopher Ferri, Marc Finzi, Micah Goldblum, Shikai Qiu, Zixi Chen","submitted_at":"2024-10-03T00:44:50Z","abstract_excerpt":"Dense linear layers are the dominant computational bottleneck in large neural networks, presenting a critical need for more efficient alternatives. Previous efforts focused on a small number of hand-crafted structured matrices and neglected to investigate whether these structures can surpass dense layers in terms of compute-optimal scaling laws when both the model size and training examples are optimally allocated. In this work, we present a unifying framework that enables searching among all linear operators expressible via an Einstein summation. This framework encompasses many previously pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.02117","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/2410.02117/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":"2410.02117","created_at":"2026-07-05T09:15:59.834322+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.02117v2","created_at":"2026-07-05T09:15:59.834322+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.02117","created_at":"2026-07-05T09:15:59.834322+00:00"},{"alias_kind":"pith_short_12","alias_value":"RFH2SRPSE5KB","created_at":"2026-07-05T09:15:59.834322+00:00"},{"alias_kind":"pith_short_16","alias_value":"RFH2SRPSE5KBZTTF","created_at":"2026-07-05T09:15:59.834322+00:00"},{"alias_kind":"pith_short_8","alias_value":"RFH2SRPS","created_at":"2026-07-05T09:15:59.834322+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.07414","citing_title":"Sparsely gated tiny linear experts","ref_index":3,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ","json":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ.json","graph_json":"https://pith.science/api/pith-number/RFH2SRPSE5KBZTTFMYXYBBI3CZ/graph.json","events_json":"https://pith.science/api/pith-number/RFH2SRPSE5KBZTTFMYXYBBI3CZ/events.json","paper":"https://pith.science/paper/RFH2SRPS"},"agent_actions":{"view_html":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ","download_json":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ.json","view_paper":"https://pith.science/paper/RFH2SRPS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.02117&json=true","fetch_graph":"https://pith.science/api/pith-number/RFH2SRPSE5KBZTTFMYXYBBI3CZ/graph.json","fetch_events":"https://pith.science/api/pith-number/RFH2SRPSE5KBZTTFMYXYBBI3CZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ/action/storage_attestation","attest_author":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ/action/author_attestation","sign_citation":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ/action/citation_signature","submit_replication":"https://pith.science/pith/RFH2SRPSE5KBZTTFMYXYBBI3CZ/action/replication_record"}},"created_at":"2026-07-05T09:15:59.834322+00:00","updated_at":"2026-07-05T09:15:59.834322+00:00"}