{"id":"W4389518993","doi":"10.18653/v1/2023.findings-emnlp.511","title":"Evaluating Dependencies in Fact Editing for Language Models: Specificity and Implication Awareness","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"Alliance de recherche numérique du Canada; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research; Samsung; Nvidia","keywords":"Dependency (UML); Computer science; Knowledge base; Protocol (science); Natural language processing; Neglect; Process (computing); Artificial intelligence; Data science; Programming language; Psychology","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01826843,0.001025081,0.0007406189,0.002836921,0.0009627219,0.003089356,0.00214274,0.002221413,0.001363376],"category_scores_gemma":[0.1224695,0.000680413,0.001216552,0.001587496,0.00126676,0.007706984,0.003506969,0.003133458,0.0003681595],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001353208,"about_ca_system_score_gemma":0.001798206,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005345766,"about_ca_topic_score_gemma":0.01037732,"domain_scores_codex":[0.9863693,0.006954496,0.0009494041,0.002714301,0.002608783,0.0004038886],"domain_scores_gemma":[0.785657,0.1844533,0.007024051,0.01706147,0.004297157,0.001506909],"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":[0.003111074,0.001979137,0.135598,0.002753484,0.00135683,0.001631604,0.008901158,0.1621458,0.03350116,0.02165159,0.01882148,0.6085487],"study_design_scores_gemma":[0.000206078,0.0006503562,0.02902578,0.0001983563,0.0004920564,0.001031318,0.001730142,0.8673914,0.03529525,0.04881203,0.01501719,0.000150069],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6585734,0.002186603,0.3165986,0.00130453,0.0001551292,0.000877654,0.004965316,0.0090419,0.006296958],"genre_scores_gemma":[0.8123714,0.0002554666,0.1788627,0.0002409854,0.00006266441,0.0002307614,0.006923364,0.0003692253,0.0006834474],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01826843,"threshold_uncertainty_score":0.09661388,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1822925834369536,"score_gpt":0.390164939008229,"score_spread":0.2078723555712754,"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."}}