{"id":"W4389520276","doi":"10.18653/v1/2023.findings-emnlp.703","title":"Investigating the Effect of Pre-finetuning BERT Models on NLI Involving Presuppositions","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Institut de Valorisation des Données; Canadian Institute for Advanced Research","keywords":"Computer science; Leverage (statistics); Presupposition; Artificial intelligence; Exploit; Machine learning; Task (project management); Theoretical computer science","routes":{"ca_aff":true,"ca_fund":true,"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":[],"consensus_categories":[],"category_scores_codex":[0.0006146798,0.000102376,0.0001220145,0.00009591176,0.0001938438,0.00007937428,0.0006916123,0.0000370175,0.000004580402],"category_scores_gemma":[0.0001517151,0.00006680571,0.00005391686,0.0004767497,0.0000360836,0.0003290034,0.0003738592,0.0001488065,0.00001603728],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001669063,"about_ca_system_score_gemma":0.00002640584,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001524219,"about_ca_topic_score_gemma":0.00001048673,"domain_scores_codex":[0.9988989,0.0001388353,0.0002150311,0.0002557392,0.0002735857,0.0002179126],"domain_scores_gemma":[0.9985039,0.000763244,0.0000657278,0.000586612,0.00003015293,0.00005037395],"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":[0.000002437319,0.00000845141,0.001146533,0.00006492516,0.00002450144,0.000002464194,0.004005,0.7231531,0.006087101,0.2515344,0.001116636,0.01285442],"study_design_scores_gemma":[0.0001137707,0.00009437585,0.0006516502,0.00008914089,0.000004706076,0.000001608297,0.00002104577,0.9787141,0.01023778,0.009985746,0.00001503935,0.00007102315],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5107377,0.00003533313,0.4782585,0.001723742,0.0001610183,0.0002878511,0.000001073694,0.0004262826,0.008368486],"genre_scores_gemma":[0.9824582,0.000002759654,0.01680261,0.0002089961,0.00004137749,0.00003351766,0.00000196984,0.000008291104,0.0004423077],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4717205,"threshold_uncertainty_score":0.2724258,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03633354126024544,"score_gpt":0.2722307729428622,"score_spread":0.2358972316826168,"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."}}