{"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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.0003161206,0.0002296731,0.0002624935,0.0002163602,0.0001875151,0.0005900117,0.001267232,0.000101793,0.0001012888],"category_scores_gemma":[0.00001596794,0.0002149874,0.00008162029,0.0008548131,0.0000218833,0.0008818263,0.0007170468,0.0001881949,0.0009754752],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001377297,"about_ca_system_score_gemma":0.00008046695,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000980016,"about_ca_topic_score_gemma":0.0001675523,"domain_scores_codex":[0.9978237,0.00005003318,0.0003324927,0.0007715587,0.0004559065,0.0005662665],"domain_scores_gemma":[0.9984077,0.000125711,0.00005950339,0.001080653,0.00006376737,0.0002626255],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0001022234,0.0002730759,0.0007584203,0.0001049965,0.0003414546,0.0007075203,0.1066381,0.263748,0.03837413,0.4495769,0.06598525,0.07338999],"study_design_scores_gemma":[0.0006360469,0.00004770585,0.0003154293,0.00004221463,0.000006306887,0.000003984478,0.000928469,0.9555786,0.0005343044,0.03782761,0.003670047,0.0004092481],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.168403,0.00004794874,0.8097077,0.001849021,0.0004546999,0.0006009,0.0000152767,0.001031724,0.0178898],"genre_scores_gemma":[0.9426216,0.00001410258,0.05249714,0.001049567,0.0002347009,0.00006061608,0.00001015975,0.00002880839,0.003483271],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7742187,"threshold_uncertainty_score":0.9998024,"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."}}