{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005908546,0.0007571623,0.0003610369,0.001271546,0.0003909056,0.001094209,0.0008885073,0.0004237787,0.009923591],"category_scores_gemma":[0.002359997,0.0003256304,0.0002499922,0.00188707,0.0002527565,0.001107527,0.000557865,0.0004742113,0.003309128],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006421559,"about_ca_system_score_gemma":0.0006791548,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002903797,"about_ca_topic_score_gemma":0.002922506,"domain_scores_codex":[0.9992431,0.00008520691,0.00007271789,0.0001901249,0.0003199948,0.00008895659],"domain_scores_gemma":[0.9984877,0.0004989589,0.0001284568,0.0003045016,0.0005035986,0.00007673044],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002469267,0.0003251734,0.01380474,0.0005463376,0.0001679377,0.0005256967,0.0004222021,0.02616298,0.1138576,0.006671849,0.04863204,0.7864142],"study_design_scores_gemma":[0.0002952536,0.0008036352,0.02279679,0.0002144422,0.0001517307,0.0006790904,0.0004663956,0.5909233,0.2870277,0.01215956,0.08435827,0.0001238117],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4717845,0.003087864,0.3451249,0.001107254,0.000312504,0.0003073661,0.004191886,0.1257685,0.04831516],"genre_scores_gemma":[0.780702,0.0009150503,0.201268,0.0002995137,0.00008112442,0.0001688946,0.004258581,0.001128924,0.01117803],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.009923591,"threshold_uncertainty_score":0.0331977,"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."}}