{"id":"W6966567684","doi":"10.48448/f9kj-h772","title":"Using Roark-Hollingshead Distance to Probe BERT's Syntactic Competence","year":2022,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Syntax; Syntactic structure; Parsing; Merge (version control); Abstract syntax tree; Predicate (mathematical logic); Simple (philosophy)","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","insufficient_payload"],"consensus_categories":["insufficient_payload"],"category_scores_codex":[0.001771123,0.0008139305,0.0008088656,0.002047546,0.00106988,0.0005066017,0.003354016,0.0001833363,0.007517569],"category_scores_gemma":[0.0006955637,0.0008257276,0.0001244327,0.00621126,0.00173125,0.0005567928,0.00129714,0.000936251,0.003186725],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002540134,"about_ca_system_score_gemma":0.002044476,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001542741,"about_ca_topic_score_gemma":0.002556299,"domain_scores_codex":[0.9921597,0.0001634413,0.0006706473,0.002294958,0.003079186,0.001632115],"domain_scores_gemma":[0.9963154,0.0001331007,0.0007577143,0.001864698,0.0002500201,0.0006791239],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"not_applicable","study_design_scores_codex":[0.0004178948,0.00311677,0.006304604,0.00136618,0.0004297481,0.001138049,0.003910213,0.03208715,0.1675531,0.1047241,0.6657223,0.01322993],"study_design_scores_gemma":[0.0005330617,0.0002502317,0.0001089433,0.0008253172,0.0001401759,0.0001430226,0.000628358,0.0335278,0.0003987216,0.0009155642,0.9604178,0.002111024],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"methods","genre_scores_codex":[0.001741904,0.002063106,0.09224001,0.001092261,0.005080219,0.006073671,0.001982345,0.003804209,0.8859223],"genre_scores_gemma":[0.0808504,0.00004734304,0.4914445,0.002233738,0.001522151,0.0003567612,0.0002566741,0.004991659,0.4182968],"genre_candidate":"other","genre_consensus":null,"teacher_disagreement_score":0.4676255,"threshold_uncertainty_score":0.9994193,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05303750460823457,"score_gpt":0.3334946302252325,"score_spread":0.2804571256169979,"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."}}