{"id":"W6891771691","doi":"10.48448/wzzn-kn70","title":"Ask LLMs Directly, \"What shapes your bias?\": Measuring Social Bias in Large Language Models","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Viewpoints; Perception; Social perception; Social change; Social comparison theory; Social influence","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":["metaepi_narrow","scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.005127575,0.0009403208,0.001002471,0.004153735,0.0003414656,0.001707213,0.002128847,0.0005952722,0.001264628],"category_scores_gemma":[0.0004959212,0.0008669451,0.0002581444,0.004676907,0.00117908,0.001584501,0.001061099,0.001242125,0.00927764],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001063126,"about_ca_system_score_gemma":0.001219425,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001735848,"about_ca_topic_score_gemma":0.01490773,"domain_scores_codex":[0.9919965,0.0002572324,0.0007897159,0.002086699,0.002933435,0.001936427],"domain_scores_gemma":[0.9981832,0.00009960795,0.0004867399,0.0007331659,0.0001655241,0.0003317698],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0002437865,0.004489902,0.001235486,0.003340488,0.0011116,0.004270181,0.1277914,0.005406454,0.05495574,0.05017506,0.5977687,0.1492112],"study_design_scores_gemma":[0.006683807,0.0003538058,0.0005254283,0.01601554,0.000822395,0.0002081445,0.06422397,0.5458283,0.005147292,0.02601067,0.3223056,0.01187507],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.0217022,0.02694058,0.001065706,0.0008290693,0.004311179,0.002394289,0.002090851,0.005575847,0.9350903],"genre_scores_gemma":[0.6322362,0.0003964051,0.002996638,0.0005044805,0.002782875,0.0001300427,0.0001871429,0.003753222,0.357013],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.610534,"threshold_uncertainty_score":0.9996483,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1335197783051541,"score_gpt":0.3466826007572805,"score_spread":0.2131628224521264,"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."}}