{"id":"W6901631708","doi":"10.60692/ayq80-vjj26","title":"DeCLUTR: Deep Contrastive Learning for Unsupervised Textual Representations","year":2021,"lang":"en","type":"article","venue":"Greater South Information System","topic":"Topic Modeling","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University Health Network; Vector Institute; University of Toronto","funders":"","keywords":"Sentence; Unsupervised learning; Metric (unit); Word (group theory); Cluster analysis; Deep learning; Limiting; Supervised learning","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":[0.001222835,0.001171148,0.0005754936,0.0005925408,0.0003883618,0.0008799676,0.00209455,0.0009977102,0.005108332],"category_scores_gemma":[0.004909346,0.0004932685,0.0007435324,0.000670595,0.0007370836,0.003071238,0.002586663,0.003052572,0.00300286],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008830854,"about_ca_system_score_gemma":0.0008081406,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001929508,"about_ca_topic_score_gemma":0.005269158,"domain_scores_codex":[0.9993456,0.0002350865,0.00003350064,0.0002280545,0.0001075038,0.00005032429],"domain_scores_gemma":[0.9988592,0.0005456547,0.00008550198,0.000283801,0.0001665803,0.00005935112],"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":[0.0003789564,0.0003619222,0.002496013,0.000427663,0.000158923,0.0002574544,0.0004210632,0.1676585,0.02935535,0.04433887,0.03715531,0.7169899],"study_design_scores_gemma":[0.00002656481,0.0001300116,0.0002767243,0.00002462515,0.00001329898,0.00007748666,0.0000338549,0.9562288,0.01015411,0.02733058,0.005684152,0.00001984212],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01213388,0.0002489152,0.9768077,0.000211516,0.00008011814,0.00007733996,0.0007440808,0.007917077,0.001779373],"genre_scores_gemma":[0.3200502,0.0003163687,0.6580778,0.0006868315,0.0001041218,0.0005109286,0.008856658,0.001320531,0.01007647],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.005108332,"threshold_uncertainty_score":0.01708907,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04129451078165032,"score_gpt":0.2396104133663395,"score_spread":0.1983159025846892,"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."}}