{"id":"W3120549770","doi":"10.18653/v1/2020.wmt-1.132","title":"NRC Systems for Low Resource German-Upper Sorbian Machine Translation 2020: Transfer Learning with Lexical Modifications","year":2020,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"National Research Council Canada","funders":"","keywords":"German; Machine translation; Transfer of learning; Computer science; Natural language processing; Artificial intelligence; Transformer; Dropout (neural networks); Linguistics; Machine learning; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.004444555,0.001920389,0.001739448,0.001661279,0.001431332,0.002545452,0.002711702,0.00179936,0.01508277],"category_scores_gemma":[0.008622997,0.000853394,0.0008617218,0.001918102,0.0006453093,0.002849205,0.003339818,0.002757284,0.02097867],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002034006,"about_ca_system_score_gemma":0.00427854,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02554713,"about_ca_topic_score_gemma":0.04764185,"domain_scores_codex":[0.9975151,0.0007337044,0.0001339625,0.0005196377,0.000800562,0.0002971009],"domain_scores_gemma":[0.9957819,0.0006656003,0.0001470612,0.001421851,0.001751789,0.0002318753],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000839346,0.0003530795,0.001567599,0.0005258727,0.0002534745,0.0003049541,0.0003174531,0.03900858,0.01687978,0.013628,0.2624987,0.6638232],"study_design_scores_gemma":[0.0003108385,0.0005166263,0.002811632,0.0001527995,0.0001704004,0.0004833882,0.0001746451,0.7389397,0.04746436,0.02147068,0.1873244,0.0001806309],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.06327147,0.004737928,0.6795691,0.002999948,0.002010494,0.001160675,0.02256748,0.1648846,0.05879827],"genre_scores_gemma":[0.333277,0.001473878,0.4819963,0.001294917,0.0004541683,0.001112913,0.1058594,0.009601834,0.06492974],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.02554713,"threshold_uncertainty_score":0.05079681,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02303102569123162,"score_gpt":0.2704460351172031,"score_spread":0.2474150094259715,"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."}}