{"id":"W38777760","doi":"10.1248/cpb.c24-00192","title":"Mots composés dans les modèles de langue pour la recherche d'information","year":2004,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Humanities; Philosophy","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":[0.001470987,0.001038125,0.0009480674,0.001041129,0.0007009532,0.003051511,0.0009146285,0.001594264,0.01081108],"category_scores_gemma":[0.004270263,0.0004892582,0.001402724,0.00130695,0.001193425,0.003321561,0.001098151,0.001819003,0.00317752],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001775183,"about_ca_system_score_gemma":0.001642785,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01363471,"about_ca_topic_score_gemma":0.01399799,"domain_scores_codex":[0.998958,0.0004314701,0.00004832395,0.0001993149,0.0003079047,0.00005487766],"domain_scores_gemma":[0.9991321,0.0005012391,0.00007575015,0.0001365447,0.0001145063,0.00003990367],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0002191335,0.00006579881,0.001516365,0.0004457237,0.0001595059,0.000508385,0.0005673533,0.2908964,0.005699825,0.597593,0.01107512,0.09125338],"study_design_scores_gemma":[0.00004233731,0.00004022267,0.0003749932,0.00008314411,0.0000475738,0.00013799,0.0001362872,0.8016073,0.002902764,0.1413079,0.05328977,0.00002973909],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0151941,0.002038805,0.9551977,0.003102217,0.0004215783,0.0001117321,0.002107693,0.002117855,0.01970834],"genre_scores_gemma":[0.4109358,0.005075687,0.5192247,0.0007181427,0.0004280026,0.001018595,0.00394412,0.0008384017,0.05781657],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01363471,"threshold_uncertainty_score":0.03616673,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1890640294505659,"score_gpt":0.3222413363651943,"score_spread":0.1331773069146284,"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."}}