{"id":"W2948710720","doi":"10.18653/v1/p19-1004","title":"Do Neural Dialog Systems Use the Conversation History Effectively? An Empirical Study","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":16,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal; Université de Montréal","funders":"Nvidia","keywords":"Dialog box; Computer science; Generative grammar; Flexibility (engineering); Shuffling; Context (archaeology); Artificial intelligence; Transformer; Code (set theory); Conversation; Dialog system; Natural language processing; Machine learning; Human–computer interaction; Programming language; Communication; World Wide Web; Psychology; Engineering","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":[0.0009605902,0.0002578934,0.0003342336,0.0001031407,0.00006442278,0.0006374512,0.001632491,0.0001848421,0.000007935036],"category_scores_gemma":[0.00007385234,0.0001733554,0.00009673369,0.00006205517,0.00003698348,0.0007045382,0.001425734,0.0006205516,0.00004134489],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006009341,"about_ca_system_score_gemma":0.0002159602,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001845449,"about_ca_topic_score_gemma":0.00004398029,"domain_scores_codex":[0.996924,0.001058731,0.0003346659,0.0009573401,0.0004980953,0.0002272006],"domain_scores_gemma":[0.9970375,0.0004113046,0.0002009293,0.002131757,0.0001388213,0.0000796852],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"observational","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00006413111,0.001281207,0.4911176,0.0003787537,0.000414367,0.00007012775,0.09811793,0.3639549,0.000107875,0.02110801,0.006979009,0.01640604],"study_design_scores_gemma":[0.0002852901,0.0001452595,0.04813758,0.00002128055,0.00002363434,0.000005000525,0.0002200176,0.9501182,0.000001484644,0.0001889708,0.0005995692,0.0002537197],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4339841,0.0001589877,0.5560324,0.000207953,0.007336735,0.001633799,0.000001099465,0.0002340124,0.0004109754],"genre_scores_gemma":[0.9976923,0.000002564391,0.0009454943,0.0002735561,0.000263264,0.0001043273,0.000003652409,0.00001482605,0.0007000571],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5861633,"threshold_uncertainty_score":0.7069228,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1060491787075014,"score_gpt":0.3092665205169398,"score_spread":0.2032173418094383,"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."}}