{"id":"W4385570971","doi":"10.18653/v1/2023.findings-acl.169","title":"From chocolate bunny to chocolate crocodile: Do Language Models Understand Noun Compounds?","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of British Columbia","funders":"Vector Institute; Universities Space Research Association; Natural Sciences and Engineering Research Council of Canada; Canadian Institute for Advanced Research","keywords":"Noun; Paraphrase; Task (project management); Interpretation (philosophy); Natural language processing; Conceptualization; Computer science; Artificial intelligence; Linguistics; Philosophy; Programming language","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001683521,0.001180475,0.0005620533,0.0007604955,0.0005034716,0.002848934,0.001503378,0.001659442,0.006320489],"category_scores_gemma":[0.01118196,0.0004161671,0.001368601,0.0008355507,0.0006068673,0.00711151,0.001546551,0.003029977,0.003097643],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001047566,"about_ca_system_score_gemma":0.001206361,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00983605,"about_ca_topic_score_gemma":0.01274702,"domain_scores_codex":[0.99915,0.0003350324,0.00004447719,0.000308027,0.00009486827,0.00006762262],"domain_scores_gemma":[0.9947569,0.00370256,0.0002554322,0.000716794,0.0004134679,0.0001548153],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.00178935,0.0006084567,0.05732978,0.002176849,0.0007963898,0.001948496,0.006484225,0.1026466,0.05900413,0.02771764,0.07546082,0.6640373],"study_design_scores_gemma":[0.0001912967,0.0001717994,0.01479908,0.0002347334,0.000153337,0.000838188,0.003623983,0.8569853,0.01578414,0.08047797,0.02663179,0.0001082885],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6177689,0.001940828,0.3208333,0.00666005,0.0005526467,0.0003640134,0.01139591,0.009364762,0.03111967],"genre_scores_gemma":[0.8780565,0.0004543948,0.1016521,0.001026051,0.00009626599,0.0001357508,0.01456534,0.0007025183,0.003311119],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.00983605,"threshold_uncertainty_score":0.02114409,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0385972335716676,"score_gpt":0.2782441317139568,"score_spread":0.2396468981422892,"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."}}