{"id":"W2584220694","doi":"10.1609/aaai.v32i1.11321","title":"RUBER: An Unsupervised Method for Automatic Evaluation of Open-Domain Dialog Systems","year":2018,"lang":"en","type":"article","venue":"Proceedings of the AAAI Conference on Artificial Intelligence","topic":"Topic Modeling","field":"Computer Science","cited_by":162,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"National Key Research and Development Program of China; National Natural Science Foundation of China; Tencent; National Science Foundation","keywords":"Dialog box; Computer science; Annotation; Metric (unit); Open domain; Artificial intelligence; Utterance; Domain (mathematical analysis); Natural language processing; Dialog system; Transferability; Conversation; Information retrieval; Machine learning; World Wide Web; Linguistics; 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":[0.01036575,0.003480121,0.00163995,0.006881806,0.001293415,0.002877639,0.003578031,0.002282785,0.006528222],"category_scores_gemma":[0.03332132,0.0007246597,0.001459436,0.002670972,0.001026263,0.004353632,0.004739637,0.002548281,0.007378978],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001384752,"about_ca_system_score_gemma":0.002317859,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003047538,"about_ca_topic_score_gemma":0.00566519,"domain_scores_codex":[0.9802033,0.009049145,0.001773899,0.004449535,0.003857481,0.0006667199],"domain_scores_gemma":[0.9841895,0.006355512,0.001165195,0.003233651,0.004440743,0.0006154997],"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":[0.0007792001,0.0004602764,0.005079574,0.001427838,0.000456553,0.0002045002,0.001746985,0.01589495,0.03504214,0.009304564,0.04821074,0.8813927],"study_design_scores_gemma":[0.0002074063,0.000726643,0.01136492,0.0002460084,0.0001695239,0.0006239208,0.001180068,0.8185634,0.0696594,0.03222704,0.06461286,0.0004188321],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01875926,0.000873649,0.9121875,0.0001559616,0.0002125812,0.0008617966,0.00425271,0.05845389,0.004242569],"genre_scores_gemma":[0.2157459,0.0002598584,0.7576846,0.0002211047,0.0001507149,0.002029165,0.01354065,0.00423133,0.006136598],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01036575,"threshold_uncertainty_score":0.05482,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.2435706272612481,"score_gpt":0.407469443211161,"score_spread":0.1638988159499128,"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."}}