{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.04434045,0.001598076,0.001120047,0.004439814,0.001030676,0.002379639,0.002036405,0.002477455,0.005042074],"category_scores_gemma":[0.1918453,0.0004057563,0.0009503541,0.00243294,0.001825235,0.004519711,0.004194188,0.001989741,0.001118754],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002721868,"about_ca_system_score_gemma":0.002365047,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003942882,"about_ca_topic_score_gemma":0.003970549,"domain_scores_codex":[0.9441975,0.04060635,0.00294766,0.0022868,0.009097489,0.0008641134],"domain_scores_gemma":[0.8010466,0.1618304,0.005275502,0.01489819,0.01490278,0.002046587],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.006461548,0.001776604,0.06080196,0.003959913,0.001751923,0.0003851747,0.003337672,0.1644019,0.01030853,0.03520485,0.05399083,0.6576191],"study_design_scores_gemma":[0.001928593,0.003865523,0.03031113,0.0009029675,0.0005531396,0.0007880732,0.001576476,0.8142278,0.02372994,0.0891562,0.03257919,0.0003809458],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"methods","genre_scores_codex":[0.5648522,0.01662337,0.3408212,0.004371254,0.001515716,0.003047808,0.007781663,0.01348428,0.04750256],"genre_scores_gemma":[0.8846503,0.0007799632,0.1013471,0.0008516063,0.0002007094,0.001370369,0.007408476,0.001179257,0.002212296],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.04434045,"threshold_uncertainty_score":0.2344975,"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."}}