{"id":"W4388555939","doi":"10.48550/arxiv.2311.04900","title":"How Abstract Is Linguistic Generalization in Large Language Models? Experiments with Argument Structure","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Northwestern University; University of Toronto; Stony Brook University; National Science Foundation; Yale University; Heinrich-Heine-Universität Düsseldorf","keywords":"Argument (complex analysis); Verb; Word order; Generalization; Linguistics; Object (grammar); Computer science; Subject (documents); Noun; Natural language processing; Space (punctuation); Artificial intelligence; Mathematics; Philosophy","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.01662033,0.002100099,0.001982182,0.0008117553,0.0007095568,0.003161466,0.002729902,0.002175791,0.002629267],"category_scores_gemma":[0.06831369,0.001115091,0.00180625,0.001081914,0.002236058,0.01510165,0.003056018,0.00544549,0.001485908],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001649185,"about_ca_system_score_gemma":0.001448204,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006025606,"about_ca_topic_score_gemma":0.009659559,"domain_scores_codex":[0.9918501,0.005040449,0.0003900356,0.001881963,0.0005688036,0.0002685335],"domain_scores_gemma":[0.9496899,0.03672923,0.001280025,0.01036483,0.001247901,0.0006881544],"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":[0.001149654,0.0003943464,0.02474994,0.000829176,0.001494405,0.0003036093,0.001932026,0.6259869,0.006567753,0.04848902,0.01437393,0.2737292],"study_design_scores_gemma":[0.000098494,0.0001861406,0.002094928,0.00009985459,0.0001353528,0.00008549313,0.000274178,0.8185283,0.002810474,0.1728449,0.002791187,0.00005068108],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2568826,0.002665256,0.7149662,0.009107914,0.0003425263,0.0002280346,0.00169709,0.006359073,0.007751297],"genre_scores_gemma":[0.8795125,0.001135176,0.1124905,0.001425942,0.0002153161,0.0002957247,0.002460736,0.0006901597,0.001773937],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01662033,"threshold_uncertainty_score":0.08789778,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07583491214773017,"score_gpt":0.2133967284531725,"score_spread":0.1375618163054423,"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."}}