{"id":"W2929767294","doi":"10.18653/v1/n19-1381","title":"Evaluating Coherence in Dialogue Systems using Entailment","year":2019,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":17,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Computer science; Coherence (philosophical gambling strategy); Textual entailment; Sentence; Natural language processing; Artificial intelligence; Logical consequence; Scalability; Quality (philosophy); Meaning (existential); Machine learning; Mathematics; Statistics; Psychology; Epistemology","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":[0.007738554,0.0008052575,0.00105653,0.002968697,0.001475936,0.00355612,0.0009932333,0.001674062,0.002556743],"category_scores_gemma":[0.05157032,0.000623418,0.0007454423,0.001521712,0.001039059,0.006287298,0.004308586,0.001338483,0.0007852116],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001232731,"about_ca_system_score_gemma":0.001377772,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003237535,"about_ca_topic_score_gemma":0.00399855,"domain_scores_codex":[0.9842626,0.01017393,0.0009885082,0.001772746,0.002352583,0.0004496603],"domain_scores_gemma":[0.964121,0.02851032,0.001798553,0.001764757,0.003225202,0.0005801031],"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":[0.008018029,0.0009941729,0.04076543,0.003779289,0.001674796,0.000943638,0.01000032,0.0959182,0.04554028,0.0571197,0.01231363,0.7229325],"study_design_scores_gemma":[0.000322102,0.00114833,0.01442381,0.0003342101,0.000606624,0.0003515216,0.003109148,0.8501715,0.02853827,0.09156652,0.009269215,0.0001587044],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5592299,0.006671589,0.4177327,0.001642907,0.0002499137,0.0007520326,0.00178512,0.002420577,0.009515191],"genre_scores_gemma":[0.9317057,0.0003578281,0.06467561,0.00005490801,0.00008756211,0.0001743646,0.002219422,0.0001244182,0.0006003049],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007738554,"threshold_uncertainty_score":0.04092592,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2056351445495016,"score_gpt":0.3773597918694956,"score_spread":0.1717246473199941,"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."}}