{"id":"W7082279606","doi":"10.48448/sc5y-vp59","title":"Social Bias Benchmark for Generation: A Comparison of Generation and QA-Based Evaluations","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Kootenay Association for Science & Technology","funders":"","keywords":"Benchmark (surveying); Measure (data warehouse); Selection bias; Adaptation (eye); Variation (astronomy); Non-response 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":[],"consensus_categories":[],"category_scores_codex":[0.0008228717,0.0001623754,0.000255306,0.000298283,0.0004188147,0.0001468728,0.0006097721,0.0001587879,0.00007888879],"category_scores_gemma":[0.0003690752,0.0001559941,0.00004649796,0.0006153604,0.0003670346,0.000109943,0.0001442885,0.00009397191,0.000001732243],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004067782,"about_ca_system_score_gemma":0.001046318,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000248191,"about_ca_topic_score_gemma":0.0001228497,"domain_scores_codex":[0.9984682,0.00004769778,0.0003145416,0.0005467485,0.0004102192,0.0002126409],"domain_scores_gemma":[0.9987784,0.00008876068,0.0003156812,0.0003323878,0.0004357759,0.0000489995],"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":[0.000005867711,0.0003708276,0.0006099803,0.000422601,0.00006230229,9.959582e-7,0.0008406871,0.008303503,0.03438957,0.1892329,0.5899057,0.175855],"study_design_scores_gemma":[0.0002944214,0.00005999325,0.00004466792,0.00003913059,0.00002414031,5.542046e-7,0.00002592579,0.9249837,0.01130471,0.001250952,0.06178727,0.0001845814],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0001135072,0.0002607372,0.8908514,0.003867479,0.0004761334,0.0004967821,0.00003147103,0.00008734296,0.1038152],"genre_scores_gemma":[0.3748741,0.00001607869,0.4159024,0.0007140323,0.001844425,0.000192479,0.000579369,0.00003326749,0.2058438],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9166802,"threshold_uncertainty_score":0.6361256,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1335914904861102,"score_gpt":0.3826334847395901,"score_spread":0.2490419942534798,"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."}}