{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0004207772,0.000713354,0.0004767812,0.0006415986,0.0005679073,0.001296952,0.00110439,0.0003742784,0.003433058],"category_scores_gemma":[0.001865477,0.0003454786,0.000543637,0.0007828138,0.000742268,0.001151334,0.001742777,0.0009931172,0.001939688],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006194778,"about_ca_system_score_gemma":0.001359766,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002002524,"about_ca_topic_score_gemma":0.002560688,"domain_scores_codex":[0.9994366,0.00008032251,0.00005018384,0.0001019983,0.0002253246,0.0001056038],"domain_scores_gemma":[0.9987023,0.0003125759,0.0001261154,0.0004819642,0.0003105397,0.000066591],"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":[0.0007455936,0.0003773792,0.004791588,0.0003068627,0.0001010453,0.00052888,0.0006532696,0.1797793,0.1960369,0.09848773,0.02428653,0.4939048],"study_design_scores_gemma":[0.00007613314,0.0001983579,0.0007380105,0.00002873299,0.00002372757,0.0001582358,0.0000589493,0.8668053,0.09441093,0.0208183,0.01663453,0.0000489203],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.08249997,0.0001547051,0.8901386,0.0002276232,0.0001573906,0.00007069929,0.0001464703,0.01868757,0.007917021],"genre_scores_gemma":[0.4619086,0.00007977261,0.5289592,0.0001718698,0.00005669974,0.0001314273,0.0006239236,0.001830802,0.006237704],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003433058,"threshold_uncertainty_score":0.01148468,"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."}}