{"id":"W6967235911","doi":"10.48448/s7em-cd36","title":"From chocolate bunny to chocolate crocodile: Do Language Models Understand Noun Compounds?","year":2023,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Religion, Ecology, and Ethics","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia; University of British Columbia Hospital","funders":"","keywords":"Conceptualization; Noun; Task (project management); Relation (database); Noun phrase; 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","sts","insufficient_payload"],"consensus_categories":["sts"],"category_scores_codex":[0.002668987,0.0004827806,0.0006462836,0.0009148757,0.001554124,0.0008607525,0.002086388,0.0008933974,0.0009092067],"category_scores_gemma":[0.0004793447,0.0004744238,0.0001328764,0.002329719,0.002745582,0.0005845433,0.0003882115,0.0007786272,0.001524877],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001003564,"about_ca_system_score_gemma":0.002217856,"about_ca_topic_candidate":true,"about_ca_topic_consensus":true,"about_ca_topic_score_codex":0.05242406,"about_ca_topic_score_gemma":0.1472648,"domain_scores_codex":[0.9948654,0.0002134038,0.0005290465,0.001339321,0.001691743,0.001361129],"domain_scores_gemma":[0.997335,0.000477924,0.0003621833,0.0008433289,0.0002613615,0.0007202059],"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":[0.00004674478,0.0002082189,0.0002845816,0.00006298893,0.0001634966,0.0001131582,0.09445779,0.001062122,0.0005106851,0.1341779,0.7662084,0.002703972],"study_design_scores_gemma":[0.0008417775,0.000172529,0.0001595127,0.0004586708,0.00008289622,0.000002319142,0.03708334,0.00274273,0.00001818551,0.08263143,0.8744329,0.001373717],"study_design_candidate":"not_applicable","study_design_consensus":"not_applicable","genre_codex":"other","genre_gemma":"other","genre_scores_codex":[0.001525255,0.0007741153,0.003284489,0.008387425,0.004430644,0.002407493,0.0003492525,0.001818083,0.9770232],"genre_scores_gemma":[0.2330286,0.007375257,0.009276425,0.004558525,0.008516002,0.0003127685,0.0001289889,0.00182377,0.7349797],"genre_candidate":"other","genre_consensus":"other","teacher_disagreement_score":0.2420435,"threshold_uncertainty_score":0.9999683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04879709904182681,"score_gpt":0.3658460927115761,"score_spread":0.3170489936697493,"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."}}