{"id":"W3006529423","doi":"10.48550/arxiv.2002.07397","title":"Improving Multi-Turn Response Selection Models with Complementary Last-Utterance Selection by Instance Weighting","year":2020,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"","keywords":"Computer science; Utterance; Task (project management); Weighting; Context (archaeology); Set (abstract data type); Selection (genetic algorithm); Artificial intelligence; Noise (video); Machine learning; Conversation; Training set; Focus (optics); Natural language processing; Data mining","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":[0.004534891,0.002030887,0.002042809,0.001240875,0.0005775611,0.00148466,0.002751664,0.00191584,0.002575274],"category_scores_gemma":[0.008240664,0.0006921109,0.001468623,0.0009147152,0.0006257563,0.002372239,0.001413944,0.002939711,0.002068362],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008586477,"about_ca_system_score_gemma":0.001016186,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005817813,"about_ca_topic_score_gemma":0.006379595,"domain_scores_codex":[0.998204,0.0009993168,0.00007002801,0.0004195655,0.0001504318,0.0001567465],"domain_scores_gemma":[0.9956005,0.003196149,0.0001598755,0.0003820805,0.0004709173,0.0001905004],"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":[0.002175401,0.0009583983,0.006868703,0.0003791636,0.0006298416,0.0002987917,0.0008812759,0.4352996,0.01865675,0.006916098,0.01562756,0.5113084],"study_design_scores_gemma":[0.00003289394,0.00005424437,0.0002957049,0.000008201961,0.00004318206,0.00002541097,0.00003844224,0.9948744,0.001602078,0.002423114,0.0005874155,0.00001498264],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1099177,0.002370925,0.8773562,0.001044973,0.0002180941,0.0003084548,0.0005343283,0.00538787,0.002861402],"genre_scores_gemma":[0.8232439,0.0004974834,0.1650812,0.0007930791,0.0003770233,0.0004961512,0.001914178,0.0005703577,0.007026574],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.005817813,"threshold_uncertainty_score":0.02398306,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.06811481835328083,"score_gpt":0.1900128017801579,"score_spread":0.1218979834268771,"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."}}