{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:HTXBVHOYV37P7NKTEPBCTPVESO","short_pith_number":"pith:HTXBVHOY","schema_version":"1.0","canonical_sha256":"3cee1a9dd8aefeffb55323c229bea493953fee87063bbc5164d02807e2f7010b","source":{"kind":"arxiv","id":"2503.01253","version":2},"attestation_state":"computed","paper":{"title":"NM-SpMM: Accelerating Matrix Multiplication Using N:M Sparsity with GPGPU","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Amelie Chi Zhou, Bingqiang Wang, Cong Ma, Du Wu, Jiang Chen, Jintao Meng, Minwen Deng, Peng Chen, Shengzhong Feng, Wenxi Zhu, Xiaowen Huang, Yanjie Wei, Yi Pan, Zhelang Deng","submitted_at":"2025-03-03T07:29:46Z","abstract_excerpt":"Deep learning demonstrates effectiveness across a wide range of tasks. However, the dense and over-parameterized nature of these models results in significant resource consumption during deployment. In response to this issue, weight pruning, particularly through N:M sparsity matrix multiplication, offers an efficient solution by transforming dense operations into semi-sparse ones. N:M sparsity provides an option for balancing performance and model accuracy, but introduces more complex programming and optimization challenges. To address these issues, we design a systematic top-down performance "},"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":"2503.01253","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.DC","submitted_at":"2025-03-03T07:29:46Z","cross_cats_sorted":[],"title_canon_sha256":"3f171ae3b1018b03b8606b1844c3ee66e0c71043d52d4d1b28e30a4a69852081","abstract_canon_sha256":"590524566f7040b571d50f9ecd097f0d408fbc3824edb76ef32a39227c8c1453"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:24:03.565137Z","signature_b64":"DDjpPgiGspUel8d2aUx0tXClD12ci70cobAkA6SKaWFtU5eIRUX6WfHSQFdqZ4VBpqjpWw8n96aMvpUn9YpqCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3cee1a9dd8aefeffb55323c229bea493953fee87063bbc5164d02807e2f7010b","last_reissued_at":"2026-07-05T10:24:03.564480Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:24:03.564480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"NM-SpMM: Accelerating Matrix Multiplication Using N:M Sparsity with GPGPU","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.DC","authors_text":"Amelie Chi Zhou, Bingqiang Wang, Cong Ma, Du Wu, Jiang Chen, Jintao Meng, Minwen Deng, Peng Chen, Shengzhong Feng, Wenxi Zhu, Xiaowen Huang, Yanjie Wei, Yi Pan, Zhelang Deng","submitted_at":"2025-03-03T07:29:46Z","abstract_excerpt":"Deep learning demonstrates effectiveness across a wide range of tasks. However, the dense and over-parameterized nature of these models results in significant resource consumption during deployment. In response to this issue, weight pruning, particularly through N:M sparsity matrix multiplication, offers an efficient solution by transforming dense operations into semi-sparse ones. N:M sparsity provides an option for balancing performance and model accuracy, but introduces more complex programming and optimization challenges. To address these issues, we design a systematic top-down performance "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.01253","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/2503.01253/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":"2503.01253","created_at":"2026-07-05T10:24:03.564563+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.01253v2","created_at":"2026-07-05T10:24:03.564563+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.01253","created_at":"2026-07-05T10:24:03.564563+00:00"},{"alias_kind":"pith_short_12","alias_value":"HTXBVHOYV37P","created_at":"2026-07-05T10:24:03.564563+00:00"},{"alias_kind":"pith_short_16","alias_value":"HTXBVHOYV37P7NKT","created_at":"2026-07-05T10:24:03.564563+00:00"},{"alias_kind":"pith_short_8","alias_value":"HTXBVHOY","created_at":"2026-07-05T10:24:03.564563+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2505.23949","citing_title":"TSENOR: Highly-Efficient Algorithm for Finding Transposable N:M Sparse Masks","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO","json":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO.json","graph_json":"https://pith.science/api/pith-number/HTXBVHOYV37P7NKTEPBCTPVESO/graph.json","events_json":"https://pith.science/api/pith-number/HTXBVHOYV37P7NKTEPBCTPVESO/events.json","paper":"https://pith.science/paper/HTXBVHOY"},"agent_actions":{"view_html":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO","download_json":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO.json","view_paper":"https://pith.science/paper/HTXBVHOY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.01253&json=true","fetch_graph":"https://pith.science/api/pith-number/HTXBVHOYV37P7NKTEPBCTPVESO/graph.json","fetch_events":"https://pith.science/api/pith-number/HTXBVHOYV37P7NKTEPBCTPVESO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO/action/storage_attestation","attest_author":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO/action/author_attestation","sign_citation":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO/action/citation_signature","submit_replication":"https://pith.science/pith/HTXBVHOYV37P7NKTEPBCTPVESO/action/replication_record"}},"created_at":"2026-07-05T10:24:03.564563+00:00","updated_at":"2026-07-05T10:24:03.564563+00:00"}