{"id":"W4412388539","doi":"10.1007/978-3-031-85240-4_14","title":"Detecting Jailbreaking Prompts: an Anti-Persuasion Filter Framework","year":2025,"lang":"en","type":"book-chapter","venue":"Lecture notes in social networks","topic":"Deception detection and forensic psychology","field":"Psychology","cited_by":2,"is_retracted":false,"has_abstract":false,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Persuasion; Computer science; Psychology; Human–computer interaction; Communication; Social psychology","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":[0.005309185,0.001288094,0.001828938,0.004111329,0.001134904,0.003790688,0.003290257,0.002998682,0.00522183],"category_scores_gemma":[0.01621273,0.0007053098,0.001099998,0.00137222,0.001475054,0.004294781,0.002049954,0.002654145,0.001384781],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001087819,"about_ca_system_score_gemma":0.001619737,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003403568,"about_ca_topic_score_gemma":0.004263005,"domain_scores_codex":[0.9979389,0.0006096224,0.0000778937,0.0004738509,0.0006418041,0.0002578891],"domain_scores_gemma":[0.9814286,0.01359959,0.00100897,0.001362415,0.002134521,0.0004659582],"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":[0.0007934019,0.00116855,0.01121332,0.0003547016,0.0002958608,0.0004158006,0.0007512831,0.08184554,0.02306467,0.1687667,0.01044361,0.7008866],"study_design_scores_gemma":[0.00001409938,0.0001038533,0.001697819,0.0000302516,0.00005669304,0.0001609071,0.00008194586,0.9350858,0.004816216,0.05606804,0.001853985,0.00003042625],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02285124,0.0003714568,0.9689191,0.0006274883,0.00008337116,0.0001406389,0.000165139,0.001226382,0.005615109],"genre_scores_gemma":[0.5310428,0.0003790161,0.4542008,0.0004221385,0.0003889349,0.0002026668,0.0004346587,0.0003043671,0.01262466],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005309185,"threshold_uncertainty_score":0.02807796,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02999823860595933,"score_gpt":0.33592303416166,"score_spread":0.3059247955557006,"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."}}