{"id":"W4416017657","doi":"10.1145/3746252.3760929","title":"Uncovering the Persuasive Fingerprint of LLMs in Jailbreaking Attacks","year":2025,"lang":"","type":"article","venue":"","topic":"Adversarial Robustness in Machine Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Guelph","funders":"","keywords":"Persuasion; Adversarial system; Readability; Fingerprint (computing); Interrogative; Sophistication","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":[0.005966012,0.0007418063,0.0005635695,0.0006731456,0.0006761837,0.00211032,0.0005858799,0.001276205,0.002477425],"category_scores_gemma":[0.06219842,0.0003763481,0.0003936189,0.0002803,0.0016874,0.002783028,0.002339226,0.003043544,0.0009355008],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004064814,"about_ca_system_score_gemma":0.000459077,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0003513803,"about_ca_topic_score_gemma":0.0006291287,"domain_scores_codex":[0.9942749,0.003783438,0.0002272556,0.0006627629,0.0007952586,0.0002563538],"domain_scores_gemma":[0.9394082,0.04545657,0.004522278,0.008025519,0.001637716,0.0009497561],"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":[0.004044222,0.001550418,0.1259678,0.002678665,0.0007294701,0.002367657,0.01747513,0.1754884,0.164692,0.0648002,0.02769772,0.4125083],"study_design_scores_gemma":[0.0001818622,0.001760999,0.03547185,0.0004938471,0.0002244867,0.001575995,0.003257838,0.7569066,0.08022308,0.09621396,0.02341114,0.0002783306],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8349394,0.0009458427,0.1470331,0.00250495,0.0002880706,0.0001804345,0.0005348187,0.004009604,0.009563806],"genre_scores_gemma":[0.9821093,0.0001046701,0.0157556,0.0003832722,0.00005173446,0.00006423163,0.0002958171,0.000272127,0.0009631852],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.005966012,"threshold_uncertainty_score":0.03155166,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01033123727741079,"score_gpt":0.2893692455782085,"score_spread":0.2790380083007977,"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."}}