{"id":"W4385573423","doi":"10.18653/v1/2022.findings-emnlp.73","title":"MCP: Self-supervised Pre-training for Personalized Chatbots with Multi-level Contrastive Sampling","year":2022,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université de Montréal","funders":"Fundamental Research Funds for the Central Universities; Ministry of Education, India; Renmin University of China; National Natural Science Foundation of China","keywords":"Computer science; Leverage (statistics); Utterance; Artificial intelligence; Dialog box; Encoder; Focus (optics); Natural language processing; Machine learning; Speech recognition; World Wide Web","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":[0.00217012,0.00149263,0.001076599,0.0006654115,0.0005300564,0.0005831676,0.002319893,0.001209675,0.002544069],"category_scores_gemma":[0.005524246,0.0006646313,0.000887374,0.0003970225,0.0007540937,0.001363073,0.001474542,0.002456013,0.001490062],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0007677386,"about_ca_system_score_gemma":0.001193684,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003262308,"about_ca_topic_score_gemma":0.006371369,"domain_scores_codex":[0.9989103,0.0004417801,0.00004116761,0.0003862833,0.0001221274,0.00009819961],"domain_scores_gemma":[0.9973817,0.001562851,0.0001392647,0.0003892963,0.0003561971,0.0001705971],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001105937,0.001049168,0.005888756,0.0004782227,0.0002306094,0.0002484894,0.0009431022,0.2727886,0.03427212,0.006182603,0.01440116,0.6624113],"study_design_scores_gemma":[0.00002303643,0.0001056456,0.0003454946,0.000008212021,0.00001232021,0.00002727477,0.00002944292,0.9934848,0.003732285,0.001489634,0.0007331878,0.000008779431],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04065141,0.0004149355,0.9498843,0.0001741068,0.00008265023,0.0002210667,0.0002162203,0.007373458,0.000981838],"genre_scores_gemma":[0.6076283,0.0001814242,0.3818691,0.0005964313,0.0001353503,0.0009401814,0.002162678,0.0005756044,0.005910907],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.003262308,"threshold_uncertainty_score":0.01147681,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1359157354290594,"score_gpt":0.3056195345867609,"score_spread":0.1697037991577015,"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."}}