{"id":"W2970009562","doi":"10.18653/v1/w19-5431","title":"Neural Machine Translation of Low-Resource and Similar Languages with Backtranslation","year":2019,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":22,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of British Columbia","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Machine translation; Nepali; Computer science; Czech; Natural language processing; Artificial intelligence; Example-based machine translation; Transformer; Transfer-based machine translation; Translation (biology); Hindi; Portuguese; Task (project management); Linguistics; Engineering","routes":{"ca_aff":true,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00009705497,0.00008126475,0.0001009931,0.00006651841,0.00001883351,0.00003705873,0.0001964852,0.00003837853,0.00002169304],"category_scores_gemma":[0.000003152097,0.00005615343,0.0000158697,0.0001677119,0.00002253229,0.0003904641,0.0000225302,0.00007411475,0.000001164657],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000003765952,"about_ca_system_score_gemma":0.000008371857,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002329411,"about_ca_topic_score_gemma":0.00001662259,"domain_scores_codex":[0.999451,0.00002209593,0.00010444,0.0001796739,0.0001526277,0.00009016274],"domain_scores_gemma":[0.9996556,0.00004595378,0.00004916226,0.0001953372,0.00003006523,0.00002388941],"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":[0.0001047279,0.00007211351,0.01224628,0.0004141008,0.00002890087,0.00001316457,0.003688803,0.0003100826,0.09822125,0.01767207,0.00008696014,0.8671415],"study_design_scores_gemma":[0.001406922,0.0005144858,0.002684112,0.0001924755,0.00001976734,0.00006028334,0.0001115924,0.7795025,0.2115763,0.002913176,0.0005090935,0.0005092398],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1701392,0.006753425,0.8164197,0.001992272,0.00002277049,0.000372334,0.00000359894,0.0005426474,0.00375405],"genre_scores_gemma":[0.7431992,0.000002646726,0.256598,0.0001058092,0.000004976436,0.000001052537,0.000003627202,0.000004486163,0.0000801866],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8666323,"threshold_uncertainty_score":0.228987,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005545113999298296,"score_gpt":0.2324726324775448,"score_spread":0.2269275184782465,"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."}}