{"id":"W4385552004","doi":"10.1017/s1351324923000396","title":"SSL-GAN-RoBERTa: A robust semi-supervised model for detecting Anti-Asian COVID-19 hate speech on social media","year":2023,"lang":"en","type":"article","venue":"Natural Language Engineering","topic":"Hate Speech and Cyberbullying Detection","field":"Computer Science","cited_by":14,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Computer science; Social media; Transformer; Artificial intelligence; Task (project management); Coronavirus disease 2019 (COVID-19); Machine learning; Speech recognition; Data mining; Voltage","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":[0.002311956,0.001719448,0.001717782,0.0008338332,0.000455607,0.0008462311,0.00315919,0.001744741,0.001455613],"category_scores_gemma":[0.00418978,0.0006216286,0.001387635,0.000531763,0.001029097,0.001371356,0.001085288,0.002725637,0.001308821],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001041063,"about_ca_system_score_gemma":0.001264279,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008012514,"about_ca_topic_score_gemma":0.01167347,"domain_scores_codex":[0.9987592,0.0005046765,0.00004899474,0.0004282052,0.0001427317,0.0001161229],"domain_scores_gemma":[0.9975339,0.001413571,0.0002039173,0.0002463929,0.0004856906,0.0001164806],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0006537822,0.0006043523,0.005909202,0.0002950033,0.000256897,0.0002801048,0.000222048,0.7308505,0.007111074,0.003889795,0.01958827,0.230339],"study_design_scores_gemma":[0.000007908388,0.00002769502,0.0001835398,0.000006154762,0.000007848043,0.0000156707,0.000007023236,0.9980325,0.0005112425,0.0009153783,0.0002783062,0.000006784386],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1216641,0.002792612,0.8560622,0.001752146,0.0004022237,0.0004246563,0.002026937,0.009575964,0.005299156],"genre_scores_gemma":[0.8490756,0.0005371608,0.1323422,0.001404941,0.0003266677,0.0005622672,0.005449168,0.0003768436,0.009925169],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008012514,"threshold_uncertainty_score":0.01593173,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0259402050914098,"score_gpt":0.260374998053483,"score_spread":0.2344347929620732,"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."}}