{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.005451299,0.0007873938,0.0004392745,0.001807052,0.0005516465,0.002990947,0.0005911145,0.0008550191,0.007412379],"category_scores_gemma":[0.1076502,0.0004334777,0.000588857,0.00152966,0.001471843,0.006881976,0.002122319,0.002083663,0.0009039871],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006772446,"about_ca_system_score_gemma":0.000680556,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004322483,"about_ca_topic_score_gemma":0.0048288,"domain_scores_codex":[0.9973662,0.00178016,0.000143744,0.0003974221,0.0002247751,0.00008763481],"domain_scores_gemma":[0.8225723,0.1617773,0.007062589,0.005153444,0.002257155,0.001177215],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.0008684993,0.0006443801,0.2088124,0.0007282707,0.0007357893,0.0005930756,0.005210773,0.2100286,0.01133894,0.1810734,0.006504571,0.3734613],"study_design_scores_gemma":[0.00002393878,0.0002038407,0.05379687,0.0001270371,0.0001305759,0.0003090323,0.000760396,0.6921911,0.004344584,0.2461945,0.001818745,0.00009943803],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6456603,0.002058174,0.317662,0.004842773,0.00007045479,0.0000991968,0.0007304091,0.001325871,0.02755085],"genre_scores_gemma":[0.9741977,0.0004644016,0.02346047,0.0001776888,0.00004352512,0.00003536259,0.000450938,0.0001050109,0.001064883],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007412379,"threshold_uncertainty_score":0.02882951,"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."}}