{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:HXAFYFGC3TLEAX3WBAY4QNIYPZ","short_pith_number":"pith:HXAFYFGC","schema_version":"1.0","canonical_sha256":"3dc05c14c2dcd6405f760831c835187e5eee2fe2d915d147fb50c4478e1d5b74","source":{"kind":"arxiv","id":"2412.11834","version":3},"attestation_state":"computed","paper":{"title":"Wonderful Matrices: Combining for a More Efficient and Effective Foundation Model Architecture","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bingheng Wu, Jingze Shi","submitted_at":"2024-12-16T14:56:28Z","abstract_excerpt":"In order to make the foundation model more efficient and effective, our idea is combining sequence transformation and state transformation. First, we prove the availability of rotary position embedding in the state space duality algorithm, which reduces the perplexity of the hybrid quadratic causal self-attention and state space duality by more than 4%, to ensure that the combining sequence transformation unifies position encoding. Second, we propose dynamic mask attention, which maintains 100% accuracy in the more challenging multi-query associative recall task, improving by more than 150% co"},"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":"2412.11834","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2024-12-16T14:56:28Z","cross_cats_sorted":["cs.AI","cs.CL"],"title_canon_sha256":"76be3c83eb09db69098209b0b0eb4efa5708d8338986c6eeadae2be497b1e37b","abstract_canon_sha256":"4e19cdc139d1f94c221ce33187e45beca0aa975e3b5e283d7140901a21f824c0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:52:19.534695Z","signature_b64":"a4ek7dTtKhQl46kObRrOtRwJUtKMdKpnWp+deCuXFtQIN7J4zolHlfoLBarHgajC+bTOKki02pM3U1qzlL/UBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3dc05c14c2dcd6405f760831c835187e5eee2fe2d915d147fb50c4478e1d5b74","last_reissued_at":"2026-07-05T09:52:19.534243Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:52:19.534243Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Wonderful Matrices: Combining for a More Efficient and Effective Foundation Model Architecture","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CL"],"primary_cat":"cs.LG","authors_text":"Bingheng Wu, Jingze Shi","submitted_at":"2024-12-16T14:56:28Z","abstract_excerpt":"In order to make the foundation model more efficient and effective, our idea is combining sequence transformation and state transformation. First, we prove the availability of rotary position embedding in the state space duality algorithm, which reduces the perplexity of the hybrid quadratic causal self-attention and state space duality by more than 4%, to ensure that the combining sequence transformation unifies position encoding. Second, we propose dynamic mask attention, which maintains 100% accuracy in the more challenging multi-query associative recall task, improving by more than 150% co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2412.11834","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/2412.11834/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":"2412.11834","created_at":"2026-07-05T09:52:19.534293+00:00"},{"alias_kind":"arxiv_version","alias_value":"2412.11834v3","created_at":"2026-07-05T09:52:19.534293+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2412.11834","created_at":"2026-07-05T09:52:19.534293+00:00"},{"alias_kind":"pith_short_12","alias_value":"HXAFYFGC3TLE","created_at":"2026-07-05T09:52:19.534293+00:00"},{"alias_kind":"pith_short_16","alias_value":"HXAFYFGC3TLEAX3W","created_at":"2026-07-05T09:52:19.534293+00:00"},{"alias_kind":"pith_short_8","alias_value":"HXAFYFGC","created_at":"2026-07-05T09:52:19.534293+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/HXAFYFGC3TLEAX3WBAY4QNIYPZ","json":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ.json","graph_json":"https://pith.science/api/pith-number/HXAFYFGC3TLEAX3WBAY4QNIYPZ/graph.json","events_json":"https://pith.science/api/pith-number/HXAFYFGC3TLEAX3WBAY4QNIYPZ/events.json","paper":"https://pith.science/paper/HXAFYFGC"},"agent_actions":{"view_html":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ","download_json":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ.json","view_paper":"https://pith.science/paper/HXAFYFGC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2412.11834&json=true","fetch_graph":"https://pith.science/api/pith-number/HXAFYFGC3TLEAX3WBAY4QNIYPZ/graph.json","fetch_events":"https://pith.science/api/pith-number/HXAFYFGC3TLEAX3WBAY4QNIYPZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ/action/storage_attestation","attest_author":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ/action/author_attestation","sign_citation":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ/action/citation_signature","submit_replication":"https://pith.science/pith/HXAFYFGC3TLEAX3WBAY4QNIYPZ/action/replication_record"}},"created_at":"2026-07-05T09:52:19.534293+00:00","updated_at":"2026-07-05T09:52:19.534293+00:00"}