{"id":"W4400382103","doi":"10.56553/popets-2024-0106","title":"Automatic Generation of Web Censorship Probe Lists","year":2024,"lang":"en","type":"article","venue":"Proceedings on Privacy Enhancing Technologies","topic":"Internet Traffic Analysis and Secure E-voting","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"National Science Foundation","keywords":"Censorship; Computer science; The Internet; Web page; Suspect; Information retrieval; Set (abstract data type); Scale (ratio); Domain (mathematical analysis); World Wide Web; Mathematics","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":[0.0005779377,0.0002193477,0.0002819598,0.0005284247,0.00008893434,0.0003966553,0.001257795,0.0001735273,0.000008874514],"category_scores_gemma":[0.0008187516,0.0001773872,0.0001201766,0.0009641316,0.00007186295,0.0005311461,0.0004233884,0.000340615,0.00004405185],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00009374746,"about_ca_system_score_gemma":0.00006012139,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000001604737,"about_ca_topic_score_gemma":0.000002815627,"domain_scores_codex":[0.9982325,0.000008723956,0.0004805035,0.0005749987,0.0003750719,0.0003282027],"domain_scores_gemma":[0.9993289,0.0000807882,0.0001699844,0.0002237593,0.0001716706,0.00002493988],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000001582518,0.00004671081,0.00004789684,0.0003760537,0.00005234855,0.000006586632,0.001830557,0.00005138348,0.06643184,0.8589434,0.001362619,0.07084899],"study_design_scores_gemma":[0.00004801503,0.0001130031,0.0000143862,0.0004432205,0.0000141141,0.00001086545,0.0001776693,0.7469666,0.2502082,0.001289847,0.0005477908,0.0001662549],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8478725,0.0005073556,0.1428659,0.002137417,0.0003146217,0.0002212948,8.284554e-7,0.005522097,0.0005579764],"genre_scores_gemma":[0.9568602,0.00003977179,0.04286075,0.00003462282,0.00005858313,0.00003607118,0.000001046099,0.00001742975,0.00009154852],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.8576536,"threshold_uncertainty_score":0.7233641,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02759634069867927,"score_gpt":0.2606456589245281,"score_spread":0.2330493182258488,"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."}}