{"id":"W3158244987","doi":"","title":"Evaluating Groundedness in Dialogue Systems: The BEGIN Benchmark","year":2021,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":15,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Benchmark (surveying); Computer science; Metric (unit); Inference; Conversation; Natural language; Artificial intelligence; Natural language processing; Machine learning; 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":[0.01238083,0.002484276,0.001302681,0.00228919,0.00127973,0.001970196,0.003114825,0.003453407,0.002896395],"category_scores_gemma":[0.04279593,0.0005201701,0.001222498,0.001304463,0.001951206,0.003496475,0.003971606,0.003286508,0.0017155],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002115436,"about_ca_system_score_gemma":0.001355016,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00467863,"about_ca_topic_score_gemma":0.007500866,"domain_scores_codex":[0.9826121,0.01132686,0.0007528585,0.002581566,0.002155505,0.0005711088],"domain_scores_gemma":[0.9634098,0.02638955,0.001388258,0.00493206,0.002565443,0.001314784],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.007614605,0.006202624,0.03260704,0.00517758,0.001908042,0.0009055543,0.003243395,0.5422639,0.02635095,0.0202043,0.08397353,0.2695486],"study_design_scores_gemma":[0.0009098647,0.003874971,0.01888197,0.000307948,0.0002342839,0.0004641002,0.001317086,0.8805325,0.03417173,0.03440341,0.02471673,0.0001853264],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6714597,0.008992088,0.2397414,0.002397466,0.0009569891,0.003259902,0.03062809,0.01710566,0.02545873],"genre_scores_gemma":[0.829583,0.0005941137,0.1162606,0.0006681909,0.0002548568,0.001848013,0.04600104,0.0007914972,0.003998626],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01238083,"threshold_uncertainty_score":0.06547683,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1780607930847837,"score_gpt":0.2358178567770648,"score_spread":0.05775706369228115,"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."}}