{"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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0005035902,0.0003306259,0.0004824408,0.001056972,0.0001697238,0.0001100265,0.0002636393,0.0003034364,0.0004198325],"category_scores_gemma":[0.0000948509,0.0003049453,0.00008784289,0.001223546,0.000406663,0.0001690675,0.0001119228,0.0005056155,0.00001127348],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001511745,"about_ca_system_score_gemma":0.0006332402,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008711532,"about_ca_topic_score_gemma":0.001165956,"domain_scores_codex":[0.997288,0.00004062824,0.0003811166,0.0007533733,0.0009248352,0.0006120324],"domain_scores_gemma":[0.9991516,0.00004589151,0.0001495115,0.0003992477,0.00004676522,0.0002069315],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007500527,0.00297632,0.00600677,0.006042666,0.0007633385,0.002416419,0.01964632,0.0395918,0.1383266,0.5446874,0.231852,0.006940189],"study_design_scores_gemma":[0.00143135,0.00008833864,0.000008834913,0.001949309,0.00007723565,0.00004497461,0.02102022,0.9676134,0.0001614427,0.006415356,0.0006573373,0.0005322172],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.003371454,0.002321715,0.460721,0.001510781,0.0005028959,0.0006988978,0.00002670895,0.0002224279,0.5306242],"genre_scores_gemma":[0.9449044,0.0004214995,0.002872741,0.0005914131,0.0001547459,0.00001011209,0.0002631041,0.0002539776,0.05052795],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.941533,"threshold_uncertainty_score":0.9999403,"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."}}