{"id":"W3156095045","doi":"10.18653/v1/2021.eacl-main.298","title":"Mega-COV: A Billion-Scale Dataset of 100+ Languages for COVID-19","year":2021,"lang":"en","type":"article","venue":"","topic":"Misinformation and Its Impacts","field":"Social Sciences","cited_by":32,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"","keywords":"Mega-; Coronavirus disease 2019 (COVID-19); Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); 2019-20 coronavirus outbreak; Scale (ratio); Computer science; Volume (thermodynamics); Virology; Geography; Medicine; Physics; Cartography; Astronomy; Outbreak","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":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0004855617,0.00004019613,0.00008650586,0.00004067227,0.0001528094,0.00004549225,0.0001105764,0.00004247217,0.003298695],"category_scores_gemma":[0.001117421,0.00003250591,0.00003635473,0.0002109643,0.00006780332,0.0002408457,0.00002333392,0.00002237653,0.00002748729],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003024228,"about_ca_system_score_gemma":0.0004658693,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001493526,"about_ca_topic_score_gemma":0.004730796,"domain_scores_codex":[0.9993955,0.00004495285,0.0001557297,0.00007366226,0.000192253,0.0001378847],"domain_scores_gemma":[0.9994131,0.0001717216,0.00005793823,0.000140354,0.00007437038,0.0001424689],"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":[0.00001126286,0.00004592965,0.0001199061,0.00005317268,0.00001110806,0.000001509486,0.02817909,0.000006995204,0.0007977643,0.04130498,0.9254892,0.003979117],"study_design_scores_gemma":[0.0002551351,0.00001357616,0.00007562715,0.000003540133,0.000005460889,8.414284e-7,0.04016183,0.0000398563,0.006471451,0.0003467595,0.9525703,0.00005566364],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.1059223,0.000581125,0.04866487,0.05186775,0.0006713124,0.001432609,0.01204623,0.000344279,0.7784695],"genre_scores_gemma":[0.8638302,0.000429797,0.01969679,0.0579699,0.0004708724,0.00001843063,0.005676104,0.00002328021,0.05188459],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7579079,"threshold_uncertainty_score":0.9976124,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08159894887163349,"score_gpt":0.438370599910293,"score_spread":0.3567716510386595,"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."}}