{"id":"W2906152891","doi":"10.1162/tacl_a_00254","title":"Analysis Methods in Neural Language Processing: A Survey","year":2019,"lang":"en","type":"article","venue":"Transactions of the Association for Computational Linguistics","topic":"Topic Modeling","field":"Computer Science","cited_by":486,"is_retracted":false,"has_abstract":true,"ca_institutions":"Artificial Intelligence in Medicine (Canada)","funders":"","keywords":"Computer science; Categorization; Artificial neural network; Field (mathematics); Feature (linguistics); Artificial intelligence; Point (geometry); Data science; Natural language processing; Machine learning; Linguistics","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":[],"consensus_categories":[],"category_scores_codex":[0.001091843,0.00007097794,0.0001853252,0.0001825871,0.00007094433,0.00004294249,0.0003973121,0.00005388455,0.000004144826],"category_scores_gemma":[0.0009145654,0.00006449757,0.0001572793,0.001030679,0.000008384871,0.00005422533,0.00001513268,0.0001088349,0.000001416203],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000135034,"about_ca_system_score_gemma":0.0001028493,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001326668,"about_ca_topic_score_gemma":0.0001184361,"domain_scores_codex":[0.9989086,0.0001844503,0.0003341014,0.0001806312,0.0002669396,0.0001252994],"domain_scores_gemma":[0.9978198,0.001099121,0.0002801667,0.0002058112,0.0005759592,0.00001911823],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00000421122,0.00004150048,0.04908028,0.00001608811,0.00007863343,3.966525e-8,0.0004784254,0.9440999,0.00001075115,0.002546921,0.000004128119,0.003639092],"study_design_scores_gemma":[0.0002306883,0.00001078579,0.095353,0.000005102137,0.00005575076,1.212935e-7,0.00001937323,0.9023499,0.00004010436,0.001797109,0.00007609629,0.00006194824],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.00855588,0.00003770179,0.9902783,0.0001318183,0.0004768416,0.0001933581,0.00004133898,0.00002921039,0.0002555696],"genre_scores_gemma":[0.7810395,2.652203e-7,0.2185252,0.00003731237,0.00002048184,0.000006312727,0.0000141941,0.000004230401,0.0003524809],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.7724836,"threshold_uncertainty_score":0.2630135,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02963772496761502,"score_gpt":0.3494461539490562,"score_spread":0.3198084289814412,"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."}}