{"id":"W6910590096","doi":"10.48448/5hfj-c153","title":"On the Relationship between Sentence Analogy Identification and Sentence Structure Encoding in Large Language Models","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Sentence; ENCODE; Analogy; Meaning (existential); Encoding (memory); Word (group theory); Identification (biology); Language model","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"],"consensus_categories":[],"category_scores_codex":[0.002071321,0.0003593381,0.000307817,0.001677686,0.0002557975,0.0003456336,0.0009219705,0.0002501346,0.0001561122],"category_scores_gemma":[0.001136834,0.0002646253,0.00004031037,0.002596421,0.0009981408,0.0003757078,0.0002956167,0.0009496525,0.0007068462],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002935305,"about_ca_system_score_gemma":0.0002314678,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0004346156,"about_ca_topic_score_gemma":0.003191358,"domain_scores_codex":[0.9966658,0.0001916707,0.0004517411,0.001127619,0.0009686057,0.0005945755],"domain_scores_gemma":[0.9980767,0.0005553992,0.0003388294,0.0008324573,0.00007522463,0.0001213365],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00001926441,0.0001234489,0.08028978,0.000414456,0.00007504141,0.0001208146,0.009129125,0.001270344,0.01593885,0.8376158,0.05390401,0.001099074],"study_design_scores_gemma":[0.001432954,0.0001623489,0.09269586,0.005186,0.0003796656,0.0001242251,0.01452056,0.3087757,0.001062113,0.5695928,0.003108486,0.002959311],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.688306,0.007383645,0.03477272,0.008871495,0.00366871,0.007132376,0.01045519,0.002853366,0.2365564],"genre_scores_gemma":[0.9713324,0.00001443372,0.0004283223,0.0001228056,0.0001750434,0.00001529182,0.0001265259,0.0002601361,0.02752505],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3075053,"threshold_uncertainty_score":0.9999806,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05434244857895414,"score_gpt":0.332230331485869,"score_spread":0.2778878829069149,"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."}}