{"id":"W3167525829","doi":"10.18653/v1/2021.naacl-main.88","title":"DReCa: A General Task Augmentation Strategy for Few-Shot Natural Language Inference","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Toyota Research Institute; Canadian Institute for Advanced Research","keywords":"Computer science; Inference; Task (project management); Natural language; Computational linguistics; Natural language processing; Artificial intelligence; Shot (pellet); Linguistics; Natural (archaeology); Engineering; Philosophy; History","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00008230973,0.0000784538,0.00008334859,0.00002853009,0.00005891896,0.000179287,0.0002664807,0.00003262725,0.00004774481],"category_scores_gemma":[0.0000339685,0.0000712725,0.00004209922,0.0001274509,0.000007582611,0.0003389327,0.0001027593,0.00006692205,0.000012255],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003164598,"about_ca_system_score_gemma":0.0001154125,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008633476,"about_ca_topic_score_gemma":0.0001614649,"domain_scores_codex":[0.9992275,0.00002593947,0.0001392157,0.0002895198,0.0001353815,0.0001824762],"domain_scores_gemma":[0.9995292,0.00005410198,0.00003198036,0.0002558913,0.00008686307,0.00004190733],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000008412075,0.00007893792,0.0002780812,0.00004974694,0.00004023526,0.00005586617,0.003354416,0.005925093,0.1357278,0.4315834,0.001559845,0.4213381],"study_design_scores_gemma":[0.0004584421,0.00003024905,0.0005580823,0.000007873121,0.000004423603,0.00001194528,0.0002486667,0.9620964,0.03317096,0.002912563,0.000335138,0.0001652781],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.152189,0.0001953965,0.844538,0.0004912221,0.000360853,0.0001029056,0.000002329159,0.00009673411,0.002023534],"genre_scores_gemma":[0.8421066,0.000003064663,0.1535345,0.0005130825,0.00009621496,0.00001712526,0.00001873122,0.000003662457,0.003707031],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9561713,"threshold_uncertainty_score":0.2906408,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0494776385202047,"score_gpt":0.3332134924141382,"score_spread":0.2837358538939335,"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."}}