{"id":"W6929291439","doi":"10.48448/pjg0-y897","title":"Understanding by Understanding Not: Modeling Negation in Language Models","year":2021,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Glaucoma and retinal disorders","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Mila - Quebec Artificial Intelligence Institute; Minnow Environmental (Canada)","funders":"","keywords":"Negation; Language model; Natural language; Core (optical fiber); Language understanding; Natural (archaeology); Data modeling","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":[0.001914411,0.0009869791,0.0005250823,0.0006859167,0.0004034241,0.00184605,0.001088756,0.001198963,0.00438081],"category_scores_gemma":[0.006003345,0.0003628025,0.001201768,0.0005042807,0.0005531982,0.003494153,0.0009953333,0.00194953,0.001847442],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008195321,"about_ca_system_score_gemma":0.001246585,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005345786,"about_ca_topic_score_gemma":0.01116648,"domain_scores_codex":[0.9991424,0.0004290506,0.00004687434,0.0002073268,0.0001219858,0.00005226583],"domain_scores_gemma":[0.9970799,0.002124051,0.0001836777,0.0002314434,0.0003000761,0.00008077617],"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":[0.0006193559,0.0004499336,0.01473639,0.001042936,0.0004479505,0.0007834244,0.001119245,0.4029985,0.01463938,0.05378784,0.05288274,0.4564923],"study_design_scores_gemma":[0.00001893144,0.00004582992,0.0005906525,0.00005187399,0.00003983581,0.0001141563,0.0001350084,0.9505306,0.00265655,0.03883217,0.006964587,0.00001978609],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1235963,0.003397634,0.8308679,0.006422262,0.000556973,0.0002548629,0.008683446,0.008307391,0.01791321],"genre_scores_gemma":[0.7545809,0.001149177,0.2193374,0.001558212,0.0002574293,0.0002189192,0.01531065,0.0006373607,0.00694998],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005345786,"threshold_uncertainty_score":0.01465529,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1135866943760007,"score_gpt":0.3172176100941626,"score_spread":0.2036309157181619,"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."}}