{"id":"W3087148478","doi":"","title":"Conditionally Adaptive Multi-Task Learning: Improving Transfer Learning in NLP Using Fewer Parameters & Less Data","year":2020,"lang":"en","type":"preprint","venue":"PolyPublie (École Polytechnique de Montréal)","topic":"Domain Adaptation and Few-Shot Learning","field":"Computer Science","cited_by":38,"is_retracted":false,"has_abstract":true,"ca_institutions":"Polytechnique Montréal","funders":"","keywords":"Multi-task learning; Computer science; Overfitting; Forgetting; Artificial intelligence; Transfer of learning; Task (project management); Machine learning; Benchmark (surveying); Transformer; Artificial neural network","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":[0.002434068,0.001620332,0.001054407,0.0006188211,0.0005158984,0.0009038537,0.003574415,0.001791705,0.003471886],"category_scores_gemma":[0.008784476,0.0005788499,0.00113187,0.0009105548,0.001017658,0.004499687,0.003398434,0.003735954,0.001342581],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009943879,"about_ca_system_score_gemma":0.001500559,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006600709,"about_ca_topic_score_gemma":0.005870451,"domain_scores_codex":[0.9991786,0.000243787,0.00004237814,0.0002801338,0.000142667,0.0001123416],"domain_scores_gemma":[0.9976104,0.00108726,0.0001490841,0.000634253,0.0003596077,0.0001592411],"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":[0.0005481652,0.0007937308,0.00238872,0.0002475418,0.000208874,0.0002891654,0.0003100916,0.4780012,0.01635172,0.006455531,0.007496583,0.4869087],"study_design_scores_gemma":[0.00002028778,0.00006799946,0.0002070238,0.00000676088,0.00001821528,0.00002435685,0.00001287306,0.9916433,0.002287626,0.005150669,0.0005513181,0.000009666686],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.089684,0.0009318038,0.8969762,0.0006014301,0.0001558805,0.000177421,0.0002497526,0.007896235,0.003327222],"genre_scores_gemma":[0.7855244,0.0004207296,0.2048533,0.0008194004,0.0001334671,0.0004357486,0.001289392,0.000592194,0.005931408],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006600709,"threshold_uncertainty_score":0.01312459,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.07018382920870495,"score_gpt":0.2793119769554278,"score_spread":0.2091281477467229,"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."}}