{"id":"W4408500266","doi":"10.1002/spe.3420","title":"Scanning HTML at Tens of Gigabytes Per Second on ARM Processors","year":2025,"lang":"en","type":"article","venue":"Software Practice and Experience","topic":"Web Data Mining and Analysis","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université TÉLUQ; Université du Québec à Montréal","funders":"Natural Sciences and Engineering Research Council of Canada","keywords":"Computer science; Operating system","routes":{"ca_aff":true,"ca_fund":true,"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":[0.0002575721,0.0001324589,0.0001872357,0.0001332673,0.0002668715,0.000143332,0.0005063334,0.00005047146,0.00005288758],"category_scores_gemma":[0.001462289,0.0001143852,0.00004153354,0.0004842012,0.0001273969,0.001214218,0.0003617927,0.0001310385,0.00001888293],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00002515556,"about_ca_system_score_gemma":0.00008648675,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00005107338,"about_ca_topic_score_gemma":0.000009393302,"domain_scores_codex":[0.9988119,0.00005048711,0.0002145362,0.0004704767,0.0002337252,0.0002188885],"domain_scores_gemma":[0.998499,0.0006343805,0.0001647052,0.0004848904,0.0001490061,0.00006796495],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0005897573,0.002101736,0.1092794,0.001130347,0.0007320694,0.0002430913,0.3140287,0.0004772094,0.02090327,0.04037512,0.02650444,0.4836348],"study_design_scores_gemma":[0.002458934,0.001726133,0.02263197,0.001907891,0.0003940293,0.0002785119,0.07101793,0.01339745,0.2285612,0.0034118,0.6516431,0.002571078],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9185526,0.001687226,0.06817478,0.002995685,0.0002812333,0.0001337887,0.00001318359,0.0002255509,0.007935922],"genre_scores_gemma":[0.9628224,0.00007963843,0.03364682,0.001410783,0.00001973039,0.00002439408,0.000002968379,0.00000547946,0.001987836],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6251386,"threshold_uncertainty_score":0.4664495,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01221089030921878,"score_gpt":0.2917902655946673,"score_spread":0.2795793752854485,"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."}}