{"id":"W4412434458","doi":"10.1016/j.comnet.2025.111513","title":"Continuous multi-task pre-training for malicious URL detection and webpage classification","year":2025,"lang":"en","type":"article","venue":"Computer Networks","topic":"Spam and Phishing Detection","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":false,"ca_institutions":"123 Certification (Canada)","funders":"National Natural Science Foundation of China","keywords":"Computer science; Task (project management); Training (meteorology); Information retrieval; Web page; Training set; Artificial intelligence; Machine learning; Data mining; World Wide Web","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":[0.0003332513,0.0001499119,0.0001807664,0.0001271483,0.0002843133,0.0003328896,0.0003320809,0.0001554793,5.005162e-7],"category_scores_gemma":[0.00001971562,0.0001560288,0.00006061985,0.0003059637,0.00003748819,0.0002655382,0.0001406275,0.0001774255,0.000001769987],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004385062,"about_ca_system_score_gemma":0.00002266171,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001643126,"about_ca_topic_score_gemma":0.00003615158,"domain_scores_codex":[0.9988713,0.00007087587,0.0002339857,0.0004775811,0.00007773555,0.0002685184],"domain_scores_gemma":[0.9991931,0.0002057425,0.0001046175,0.0003522159,0.0000850811,0.00005928029],"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":[0.00002335069,0.00003080652,0.0003577436,0.00002287129,0.00003140841,0.000001398402,0.0006251941,0.004047966,0.0007782663,0.002645758,0.0008936233,0.9905416],"study_design_scores_gemma":[0.0005883535,0.00009655494,0.02815956,0.00004894521,0.00001312821,0.00001272408,0.0000109463,0.9626567,0.0001672637,0.0007508578,0.007344634,0.0001503275],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0171475,0.0003635501,0.9787637,0.0002392888,0.002609497,0.0004207742,8.752496e-7,0.0003622838,0.00009254264],"genre_scores_gemma":[0.9454301,0.00002339623,0.05334393,0.0004820734,0.0004480595,0.0000734338,0.000004647582,0.00001036095,0.0001839891],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9903913,"threshold_uncertainty_score":0.6362671,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02498762403068101,"score_gpt":0.2638351869955468,"score_spread":0.2388475629648658,"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."}}