{"id":"W6928872153","doi":"10.48448/xqtn-hn45","title":"Everyone's Voice Matters: Quantifying Annotation Disagreement Using Demographic Information","year":2023,"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":"Annotation; Task (project management); Variance (accounting); Demographics; Process (computing); Ground truth","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","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001749122,0.0005007632,0.0003823887,0.00448285,0.0005089433,0.0006180176,0.0009225429,0.0002323073,0.0002554937],"category_scores_gemma":[0.0002687558,0.0004974993,0.0001005752,0.005645844,0.0009922757,0.002939957,0.000306536,0.0003477799,0.009453858],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007001613,"about_ca_system_score_gemma":0.0005009143,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002362669,"about_ca_topic_score_gemma":0.003398732,"domain_scores_codex":[0.99545,0.00008144933,0.0007370039,0.0006855124,0.002159158,0.0008868928],"domain_scores_gemma":[0.9976334,0.00008963515,0.001034565,0.0007388447,0.000294582,0.0002089571],"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":[0.00009451231,0.0004925464,0.02296756,0.001558643,0.0005173805,0.00005012606,0.00439695,0.020482,0.02585351,0.004257053,0.8752478,0.04408193],"study_design_scores_gemma":[0.002097303,0.0002100792,0.01150041,0.003872419,0.0004855014,0.00006066015,0.00406216,0.5438164,0.0004972876,0.0005421124,0.4293223,0.00353337],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.05577553,0.003419859,0.5710688,0.005301625,0.04211921,0.02240797,0.006899959,0.05364903,0.239358],"genre_scores_gemma":[0.5409318,0.001039759,0.2703194,0.01230683,0.006416222,0.000625417,0.00652496,0.02081929,0.1410163],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5233344,"threshold_uncertainty_score":0.9997477,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06876708129350656,"score_gpt":0.3409416048049575,"score_spread":0.2721745235114509,"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."}}