{"id":"W6966841497","doi":"10.48448/2s2c-3s80","title":"Context-aware Adversarial Training for Name Regularity Bias in Named Entity Recognition","year":2021,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"University of Waterloo; Université de Montréal","funders":"","keywords":"Adversarial system; Training (meteorology); Named-entity recognition; Training set; Pattern recognition (psychology); Key (lock)","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"],"consensus_categories":[],"category_scores_codex":[0.001825346,0.0002993882,0.0004709559,0.0006998998,0.0002010484,0.0003568107,0.001614591,0.0003187562,0.0002286373],"category_scores_gemma":[0.0007621422,0.000317812,0.0001153711,0.0009374579,0.0003577179,0.0004615671,0.0004387203,0.0003942238,0.0000169966],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002967664,"about_ca_system_score_gemma":0.0014892,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008334349,"about_ca_topic_score_gemma":0.003908247,"domain_scores_codex":[0.9967029,0.0001133354,0.0004785233,0.001310731,0.0007378781,0.0006566459],"domain_scores_gemma":[0.9982735,0.000170819,0.0003244695,0.0008369936,0.0002309254,0.0001632754],"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":[0.00002159588,0.0002528305,0.0002731191,0.0002745512,0.00005062871,0.00007611133,0.002797592,0.0002203238,0.0004331557,0.02975842,0.008179128,0.9576625],"study_design_scores_gemma":[0.002933286,0.0001469807,0.0001379399,0.001523545,0.00003278704,0.0000405791,0.001003233,0.8795824,0.0004946034,0.02348481,0.08928744,0.001332372],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0007310672,0.0003251023,0.9762163,0.0009289318,0.003296752,0.0008351972,0.00005709646,0.0003071638,0.01730242],"genre_scores_gemma":[0.2807709,0.0001112546,0.6596089,0.00167581,0.002253906,0.0002225056,0.0004172758,0.000307938,0.05463153],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9563302,"threshold_uncertainty_score":0.9999274,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1559789922302824,"score_gpt":0.3242821735167069,"score_spread":0.1683031812864245,"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."}}