{"id":"W2963172394","doi":"10.1109/icassp.2019.8682634","title":"Why Do Neural Dialog Systems Generate Short and Meaningless Replies? a Comparison between Dialog and Translation","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":12,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Natural Science Foundation of China","keywords":"Dialog box; Computer science; Utterance; Natural language processing; Machine translation; Randomness; Dialog system; Sequence (biology); Artificial intelligence; Translation (biology); Speech recognition; Conjecture; World Wide Web; Mathematics","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":[0.007172694,0.0008076675,0.000906872,0.000888437,0.0009077706,0.001630367,0.0009923682,0.001778667,0.003880687],"category_scores_gemma":[0.04590055,0.000438076,0.0006126781,0.0007011016,0.001325341,0.003955623,0.001326355,0.001295177,0.001765287],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008236902,"about_ca_system_score_gemma":0.0008139446,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001472484,"about_ca_topic_score_gemma":0.001378103,"domain_scores_codex":[0.9954608,0.002771712,0.0002449848,0.0008538399,0.0004896629,0.0001791057],"domain_scores_gemma":[0.9696474,0.02384852,0.001053208,0.00292999,0.002022682,0.0004980579],"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":[0.00544208,0.0008905565,0.03483778,0.003211199,0.0009094533,0.001050729,0.006537287,0.2426362,0.06897146,0.05538908,0.01167832,0.5684458],"study_design_scores_gemma":[0.000381205,0.001182485,0.02114985,0.0001771631,0.0002725059,0.0009091556,0.001535168,0.8134978,0.04158442,0.1087485,0.01034159,0.0002201657],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5847161,0.003518136,0.3823633,0.003730534,0.0003661306,0.0003061216,0.00133719,0.003538664,0.02012373],"genre_scores_gemma":[0.9666745,0.0003220973,0.02981027,0.0004036989,0.00008575751,0.0001511181,0.0008651398,0.0001653341,0.00152218],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007172694,"threshold_uncertainty_score":0.03793323,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05761809124299732,"score_gpt":0.2660111035770797,"score_spread":0.2083930123340824,"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."}}