{"id":"W6947874636","doi":"10.48448/xwge-6p50","title":"A Closer Look at Linguistic Knowledge in Masked Language Models: The Case of Relative Clauses in American English","year":2020,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Amphibian and Reptile Biology","field":"Environmental Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Mila - Quebec Artificial Intelligence Institute","funders":"","keywords":"Antecedent (behavioral psychology); Focus (optics); Task (project management); Identification (biology); Semantics (computer science); Passive voice; Strengths and weaknesses; Computational linguistics","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":["sts","insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000423842,0.000225542,0.0003559442,0.0002145782,0.00006943814,0.0000121225,0.0006173597,0.0001305909,0.0009367975],"category_scores_gemma":[0.000346092,0.0001525334,0.00004741957,0.001192045,0.00436066,0.00006695939,0.0006424721,0.0003498693,0.0001488546],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0003723628,"about_ca_system_score_gemma":0.0001204484,"about_ca_topic_candidate":true,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006382894,"about_ca_topic_score_gemma":0.07370469,"domain_scores_codex":[0.9983314,0.0001480236,0.0003230936,0.000625283,0.0001973314,0.0003749205],"domain_scores_gemma":[0.9990618,0.000164451,0.0002592631,0.0003965092,0.00001299535,0.0001049554],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"qualitative","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0005138206,0.003594295,0.1133747,0.0005772124,0.0002432943,0.01538592,0.5367312,0.01121362,0.0127617,0.04133144,0.133058,0.1312147],"study_design_scores_gemma":[0.007350372,0.003042146,0.01448225,0.002017201,0.0003114509,0.0008483773,0.1171346,0.3864079,0.001765495,0.0102758,0.4503314,0.006032999],"study_design_candidate":"qualitative","study_design_consensus":null,"genre_codex":"other","genre_gemma":"empirical","genre_scores_codex":[0.1167707,0.0008066164,0.0001542305,0.0001794205,0.0004315111,0.0007742535,0.00007001464,0.00009196351,0.8807213],"genre_scores_gemma":[0.9645678,0.00007812746,0.001162302,0.0001731654,0.000143682,0.00001415876,0.000008936461,0.00006872872,0.03378309],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8477971,"threshold_uncertainty_score":0.9999765,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01577317363147352,"score_gpt":0.2736332674111671,"score_spread":0.2578600937796935,"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."}}