{"id":"W4389520434","doi":"10.18653/v1/2023.findings-emnlp.567","title":"Contrastive Deterministic Autoencoders For Language Modeling","year":2023,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Vector Institute; University of Waterloo","funders":"","keywords":"Computer science; Language model; Artificial intelligence; Transformer; Mixture model; Machine learning; Representation (politics); Natural language processing","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001894181,0.00007467941,0.00009554865,0.00007505874,0.00007792222,0.00007090419,0.0003814163,0.00003055938,0.000005351448],"category_scores_gemma":[0.00009152523,0.00006745414,0.00004689912,0.0001604898,0.000008743724,0.000155751,0.0001018809,0.00004192773,0.00006164012],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001754857,"about_ca_system_score_gemma":0.00004271613,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002473866,"about_ca_topic_score_gemma":0.00001100504,"domain_scores_codex":[0.9992116,0.00001099031,0.0001363261,0.0002690971,0.0001115768,0.0002604184],"domain_scores_gemma":[0.9994836,0.0001437146,0.00001999238,0.0002614569,0.00003771807,0.00005353073],"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":[0.000009943938,0.0000246824,0.0000558082,0.0000568563,0.00003503914,0.00005992239,0.007438917,0.6610926,0.001531869,0.2163268,0.001192457,0.1121751],"study_design_scores_gemma":[0.0001968963,0.00001920344,0.00001103048,0.000006952533,0.000002651296,0.000002306456,0.0002630953,0.9929995,0.0001269936,0.006217459,0.00006244177,0.00009150148],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01424872,0.00001287766,0.9825851,0.0004407255,0.0002554172,0.000169619,0.000002334263,0.000656484,0.001628744],"genre_scores_gemma":[0.8496323,0.000001267155,0.14913,0.0002643052,0.00005314677,0.00003752578,0.000002287768,0.000007012302,0.0008722235],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8353835,"threshold_uncertainty_score":0.27507,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0410131605871742,"score_gpt":0.2989528435363168,"score_spread":0.2579396829491427,"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."}}