{"id":"W4376562663","doi":"10.1002/spe.3214","title":"Fast matrix multiplication via compiler‐only layered data reorganization and intrinsic lowering","year":2023,"lang":"en","type":"article","venue":"Software Practice and Experience","topic":"Parallel Computing and Optimization Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"IBM (Canada); University of Alberta","funders":"","keywords":"Computer science; Compiler; Parallel computing; Kernel (algebra); Matrix multiplication; Supercomputer; Code (set theory); Performance improvement; Matrix (chemical analysis); Code generation; Computational science; Programming language; Operating system","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003716715,0.0001229174,0.0001169513,0.000108496,0.0003200151,0.0003158866,0.0006062347,0.00005993914,0.000002631301],"category_scores_gemma":[0.001009408,0.0001273743,0.000008293462,0.0007482428,0.00006232748,0.002271301,0.001084193,0.000117092,0.00002692265],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000016376,"about_ca_system_score_gemma":0.00004043265,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004973511,"about_ca_topic_score_gemma":0.000002449776,"domain_scores_codex":[0.9987779,0.00006324304,0.0002099701,0.0005631557,0.000190466,0.0001952818],"domain_scores_gemma":[0.998541,0.00034745,0.0001543356,0.0007268653,0.0001473191,0.00008301503],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00007818823,0.0002077553,0.02179214,0.0001440717,0.00004773194,0.00007172662,0.03791413,0.003292273,0.003816297,0.00534974,0.004883035,0.9224029],"study_design_scores_gemma":[0.000714741,0.0001686204,0.01412932,0.0001129281,0.00002154418,0.0004532394,0.001665793,0.9520705,0.003031453,0.001266816,0.02553218,0.0008328153],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02155393,0.0002381261,0.9756719,0.0009794682,0.000119086,0.0001539001,0.000003333899,0.00121993,0.00006030569],"genre_scores_gemma":[0.7006536,0.0007284222,0.2981601,0.0002849701,0.00004303411,0.00001255094,0.00004497944,0.00001215351,0.00006019501],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9487783,"threshold_uncertainty_score":0.5194174,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02645396992790152,"score_gpt":0.3220810229898587,"score_spread":0.2956270530619572,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}