{"id":"W4384659592","doi":"10.1145/3539618.3592007","title":"Learning from Crowds with Annotation Reliability","year":2023,"lang":"en","type":"article","venue":"","topic":"Mobile Crowdsensing and Crowdsourcing","field":"Computer Science","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"National Key Research and Development Program of China; Canadian Institute for Advanced Research","keywords":"Crowds; Computer science; Annotation; Crowdsourcing; Reliability (semiconductor); Quality (philosophy); Artificial intelligence; Machine learning; Supervised learning; World Wide Web","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003207087,0.00008967226,0.00009632608,0.00006720671,0.0001929169,0.0001983246,0.0002332584,0.00003856682,0.00002053563],"category_scores_gemma":[0.00008474978,0.00007016402,0.00002753926,0.0006736945,0.00003222519,0.0002912834,0.00009727402,0.0001400042,0.000280664],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0000214299,"about_ca_system_score_gemma":0.00003460067,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002501111,"about_ca_topic_score_gemma":0.00001906464,"domain_scores_codex":[0.999003,0.00006833488,0.0001244569,0.0003703748,0.000217989,0.0002158725],"domain_scores_gemma":[0.9992254,0.0001835677,0.0000425377,0.0004130454,0.00007551535,0.00005996146],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006775442,0.0001128707,0.2023944,0.00005952055,0.00008087448,0.0002366072,0.01625677,0.2897452,0.024645,0.01959341,0.0126909,0.4341168],"study_design_scores_gemma":[0.0004340093,0.0001934799,0.1280859,0.0000465942,0.000007141731,0.000007941165,0.0004895782,0.852449,0.008437084,0.003169894,0.006331007,0.0003483648],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6969085,0.000004828796,0.2971094,0.0006141666,0.0001045564,0.00005196186,2.312963e-7,0.00110083,0.004105542],"genre_scores_gemma":[0.9796847,0.000002228504,0.01830596,0.00009935888,0.00004575281,0.000005750864,0.000006764598,0.00000834758,0.001841113],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.5627038,"threshold_uncertainty_score":0.3607461,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01142408964575587,"score_gpt":0.2164370281660434,"score_spread":0.2050129385202875,"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."}}