{"id":"W4402888402","doi":"10.2196/60678","title":"Evaluating the Influence of Role-Playing Prompts on ChatGPT’s Misinformation Detection Accuracy: Quantitative Study","year":2024,"lang":"en","type":"article","venue":"JMIR Infodemiology","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":11,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Misinformation; Disinformation; Psychology; Social media; Social psychology; Identity (music); Social identity theory; Applied psychology; Political science; Computer science; Computer security; Social group","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.003686121,0.0001140329,0.0001717793,0.0001745347,0.0004710277,0.0000946787,0.0002376787,0.0001175064,0.00004038968],"category_scores_gemma":[0.004286547,0.0000784119,0.00005330593,0.0004434243,0.0001718631,0.0009762534,0.00004967141,0.0002940479,0.0001167821],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001136023,"about_ca_system_score_gemma":0.0002291001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004970035,"about_ca_topic_score_gemma":0.0002244097,"domain_scores_codex":[0.9980884,0.0005698868,0.0005521013,0.0001353782,0.000399733,0.0002545217],"domain_scores_gemma":[0.9977559,0.001500856,0.0003049939,0.0001856726,0.0001973915,0.00005515285],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.0001893511,0.00007490334,0.002713217,0.00007394181,0.0000559078,7.018787e-7,0.8439966,0.02970894,0.003683735,0.0407493,0.0001488422,0.0786046],"study_design_scores_gemma":[0.0008652908,0.003688198,0.5755806,0.0003405353,0.0000583942,0.000007785035,0.2169847,0.1920435,0.0006071334,0.005694331,0.003668338,0.000461143],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9904019,0.00003875602,0.0001991195,0.0006550879,0.0002298409,0.001187363,0.000003223685,0.0001189896,0.00716577],"genre_scores_gemma":[0.9992852,0.00001669032,0.0001227767,0.0003967344,0.00005437216,0.0000587796,0.000002662549,0.000005333101,0.00005746552],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.6270119,"threshold_uncertainty_score":0.5131707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1340513233802499,"score_gpt":0.4954393638931831,"score_spread":0.3613880405129332,"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."}}