{"id":"W2125247927","doi":"10.3115/1690219.1690223","title":"Summarizing multiple spoken documents","year":2009,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":40,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Computer science; Salient; Transcription (linguistics); Speech recognition; Vocabulary; Word (group theory); Spoken language; Word error rate; Artificial intelligence; Natural language processing; Feature (linguistics); Language model; Independence (probability theory); 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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00126939,0.001134176,0.001050919,0.001717185,0.0006105792,0.002241894,0.001736914,0.001218541,0.005839161],"category_scores_gemma":[0.006477901,0.00049998,0.001367354,0.00143573,0.0004714956,0.002812332,0.001267852,0.001210769,0.002999827],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007014634,"about_ca_system_score_gemma":0.0009807845,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002911658,"about_ca_topic_score_gemma":0.003143453,"domain_scores_codex":[0.9983764,0.0003522169,0.0001504245,0.000547856,0.0004853543,0.00008774825],"domain_scores_gemma":[0.9971988,0.00132622,0.0002854567,0.0003989353,0.0006978023,0.00009276617],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0007502634,0.0002190599,0.004138961,0.001366275,0.0004460757,0.001362363,0.002899819,0.09370811,0.0406713,0.04306391,0.01930744,0.7920664],"study_design_scores_gemma":[0.0001341814,0.0005500226,0.004151208,0.0002257443,0.0006017587,0.001331495,0.001255793,0.8289921,0.03019569,0.0715449,0.06075857,0.0002584536],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03262134,0.001681655,0.9533317,0.0009401205,0.0005248461,0.0002397864,0.002657754,0.003370906,0.004631921],"genre_scores_gemma":[0.424043,0.001934103,0.5463611,0.0004737415,0.000901295,0.0005692203,0.006853488,0.0006204777,0.0182436],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005839161,"threshold_uncertainty_score":0.01953399,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01945310161423677,"score_gpt":0.2523597196129386,"score_spread":0.2329066179987018,"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."}}