{"id":"W6948968340","doi":"10.5281/zenodo.10428713","title":"Large Language Model (LLM) Bias Index—LLMBI","year":2023,"lang":"en","type":"other","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Libraries and Information Services","field":"Arts and Humanities","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Canada West","funders":"","keywords":"Measure (data warehouse); Reliability (semiconductor); Empirical measure; Language model; Process (computing); Diversity (politics); Gender bias","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":["sts","scholarly_communication","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.0003001082,0.0002130314,0.0002098105,0.0004355878,0.001489108,0.002113675,0.0009413201,0.0001196944,0.2149543],"category_scores_gemma":[0.0001009348,0.0002004624,0.00007780989,0.000162525,0.0001320767,0.0005325185,0.001016231,0.0002779976,0.0475065],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004669106,"about_ca_system_score_gemma":0.000005002826,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000175094,"about_ca_topic_score_gemma":0.00005634326,"domain_scores_codex":[0.9985837,0.00008853043,0.000274967,0.0002860981,0.0003884874,0.0003781681],"domain_scores_gemma":[0.9989942,0.00001384944,0.0002157526,0.0004692209,0.0001870977,0.0001198407],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.000008792156,0.00004263285,1.749091e-7,0.0001134584,0.00005015247,0.000003916717,0.01713848,0.00001373214,0.000001895508,0.1038538,0.8734298,0.005343163],"study_design_scores_gemma":[0.0003092968,0.00004911993,0.00000266751,0.00007410392,0.00001263846,0.000003494327,0.00576161,0.00207788,0.000003977831,0.00019174,0.9912714,0.0002420587],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.00007504033,0.0001382351,0.0002684108,0.000319413,0.0002680029,0.0003096894,0.002320388,0.00415315,0.9921477],"genre_scores_gemma":[0.007529324,0.0002432243,0.0000304466,0.000699927,0.001073231,9.715118e-8,0.005110554,0.01315047,0.9721627],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.1674478,"threshold_uncertainty_score":0.9998108,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0665843376175713,"score_gpt":0.2379908573496372,"score_spread":0.1714065197320659,"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."}}