{"id":"W4388757737","doi":"10.1109/uemcon59035.2023.10316060","title":"Intentional Biases in LLM Responses","year":2023,"lang":"en","type":"article","venue":"","topic":"Persona Design and Applications","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Victoria","funders":"","keywords":"Viewpoints; Variety (cybernetics); Persona; Computer science; Construct (python library); Set (abstract data type); Language model; Field (mathematics); Supervisor; Human–computer interaction; Artificial intelligence; Data science; Natural language processing; Programming language","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":[0.02806496,0.0008249049,0.0004278987,0.00105194,0.001514801,0.003649292,0.0008325081,0.001564017,0.006858917],"category_scores_gemma":[0.1834443,0.0004002166,0.0004535156,0.0004819246,0.002737446,0.003949361,0.005768519,0.002369422,0.002703801],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001563629,"about_ca_system_score_gemma":0.0007525979,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0005825981,"about_ca_topic_score_gemma":0.000505425,"domain_scores_codex":[0.9157214,0.06845661,0.002813113,0.003744859,0.007596623,0.001667374],"domain_scores_gemma":[0.8749535,0.08926962,0.008331139,0.01683692,0.00897492,0.00163386],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"observational","study_design_scores_codex":[0.002335436,0.0004677537,0.098933,0.001422024,0.0001953398,0.000986648,0.4681196,0.003339519,0.08736856,0.08457536,0.01307763,0.2391792],"study_design_scores_gemma":[0.0004710759,0.002013191,0.07952435,0.002130545,0.00039641,0.003408891,0.2804622,0.05248296,0.08227754,0.1675643,0.3283878,0.000880675],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.7534143,0.0003868545,0.1619568,0.004907944,0.0004648248,0.0005796076,0.0003456526,0.001782817,0.07616118],"genre_scores_gemma":[0.9810382,0.00006937652,0.01264645,0.001477906,0.00005577634,0.0004039831,0.0001295961,0.000326434,0.003852249],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02806496,"threshold_uncertainty_score":0.1484234,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1068782892838309,"score_gpt":0.3323295776576057,"score_spread":0.2254512883737749,"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."}}