{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:F46VGQI4IX5ABD3HWAAF7EZMFE","short_pith_number":"pith:F46VGQI4","schema_version":"1.0","canonical_sha256":"2f3d53411c45fa008f67b0005f932c292f8c816eed7200769f5086d6b9c7cce9","source":{"kind":"arxiv","id":"2409.13313","version":1},"attestation_state":"computed","paper":{"title":"Performance Enhancement of the Ozaki Scheme on Integer Matrix Multiplication Unit","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Katsuhisa Ozaki, Toshiyuki Imamura, Yuki Uchino","submitted_at":"2024-09-20T08:15:42Z","abstract_excerpt":"This study was aimed at simultaneously achieving sufficient accuracy and high performance for general matrix multiplications. Recent architectures, such as NVIDIA GPUs, feature high-performance units designed for low-precision matrix multiplications in machine learning models, and next-generation architectures are expected to follow the same design principle. The key to achieving superior performance is to fully leverage such architectures. The Ozaki scheme, a highly accurate matrix multiplication algorithm using error-free transformations, enables higher-precision matrix multiplication to be "},"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":"2409.13313","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2024-09-20T08:15:42Z","cross_cats_sorted":[],"title_canon_sha256":"968aaad9e61df03499c9b78bdb481e7dc75bab858e2162c8429cfecb1c1fd40f","abstract_canon_sha256":"519fd02431533b5f88957ac43ba7230719d539835ce806dbe69c1f8d7d2c4fbb"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:54:36.598335Z","signature_b64":"jD3j39lhJ7Bf1Dr2n+6WjP0FhaKZukdtwZtZJyHuE+Gt0Vg5IgXKbqJj6Q2eAQoXu+gemk3cOM3viU94OlkpAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f3d53411c45fa008f67b0005f932c292f8c816eed7200769f5086d6b9c7cce9","last_reissued_at":"2026-07-05T10:54:36.597826Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:54:36.597826Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Performance Enhancement of the Ozaki Scheme on Integer Matrix Multiplication Unit","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Katsuhisa Ozaki, Toshiyuki Imamura, Yuki Uchino","submitted_at":"2024-09-20T08:15:42Z","abstract_excerpt":"This study was aimed at simultaneously achieving sufficient accuracy and high performance for general matrix multiplications. Recent architectures, such as NVIDIA GPUs, feature high-performance units designed for low-precision matrix multiplications in machine learning models, and next-generation architectures are expected to follow the same design principle. The key to achieving superior performance is to fully leverage such architectures. The Ozaki scheme, a highly accurate matrix multiplication algorithm using error-free transformations, enables higher-precision matrix multiplication to be "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2409.13313","kind":"arxiv","version":1},"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/2409.13313/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":"2409.13313","created_at":"2026-07-05T10:54:36.597887+00:00"},{"alias_kind":"arxiv_version","alias_value":"2409.13313v1","created_at":"2026-07-05T10:54:36.597887+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2409.13313","created_at":"2026-07-05T10:54:36.597887+00:00"},{"alias_kind":"pith_short_12","alias_value":"F46VGQI4IX5A","created_at":"2026-07-05T10:54:36.597887+00:00"},{"alias_kind":"pith_short_16","alias_value":"F46VGQI4IX5ABD3H","created_at":"2026-07-05T10:54:36.597887+00:00"},{"alias_kind":"pith_short_8","alias_value":"F46VGQI4","created_at":"2026-07-05T10:54:36.597887+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.11277","citing_title":"Analysis of Floating-Point Matrix Multiplication Computed via Integer Arithmetic","ref_index":47,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE","json":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE.json","graph_json":"https://pith.science/api/pith-number/F46VGQI4IX5ABD3HWAAF7EZMFE/graph.json","events_json":"https://pith.science/api/pith-number/F46VGQI4IX5ABD3HWAAF7EZMFE/events.json","paper":"https://pith.science/paper/F46VGQI4"},"agent_actions":{"view_html":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE","download_json":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE.json","view_paper":"https://pith.science/paper/F46VGQI4","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2409.13313&json=true","fetch_graph":"https://pith.science/api/pith-number/F46VGQI4IX5ABD3HWAAF7EZMFE/graph.json","fetch_events":"https://pith.science/api/pith-number/F46VGQI4IX5ABD3HWAAF7EZMFE/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE/action/storage_attestation","attest_author":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE/action/author_attestation","sign_citation":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE/action/citation_signature","submit_replication":"https://pith.science/pith/F46VGQI4IX5ABD3HWAAF7EZMFE/action/replication_record"}},"created_at":"2026-07-05T10:54:36.597887+00:00","updated_at":"2026-07-05T10:54:36.597887+00:00"}