{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001085201,0.0007026681,0.0005115901,0.0004195419,0.0003387975,0.0009545879,0.001112649,0.0008656229,0.003640421],"category_scores_gemma":[0.00503181,0.0005333641,0.0009121317,0.0005275885,0.0007495468,0.001684689,0.00101605,0.002578242,0.001320699],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001018374,"about_ca_system_score_gemma":0.0007689331,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004114348,"about_ca_topic_score_gemma":0.007569149,"domain_scores_codex":[0.9995824,0.0001772597,0.00002134278,0.0001012954,0.00008474093,0.00003306151],"domain_scores_gemma":[0.9983919,0.001193909,0.00007635955,0.0001814907,0.0001221938,0.0000340629],"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":[0.00006292897,0.00003779991,0.0005168185,0.0001084727,0.00006761358,0.00007688506,0.0001070621,0.808593,0.005439093,0.1094464,0.003038182,0.07250577],"study_design_scores_gemma":[0.000002008083,0.000006949197,0.00004748362,0.000005689056,0.000003571611,0.000012291,0.000003369211,0.9771789,0.0007347897,0.02116722,0.0008335125,0.000004173614],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.006368266,0.0004106976,0.9902866,0.000261824,0.00004094518,0.0000150488,0.0001250889,0.0005950877,0.001896404],"genre_scores_gemma":[0.573767,0.001119583,0.4128282,0.0004122526,0.0001705836,0.0002002878,0.0008674336,0.0005528141,0.01008191],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004114348,"threshold_uncertainty_score":0.01217842,"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."}}