{"id":"W3116587780","doi":"10.18653/v1/2020.coling-main.410","title":"Revitalization of Indigenous Languages through Pre-processing and Neural Machine Translation: The case of Inuktitut","year":2020,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":19,"is_retracted":false,"has_abstract":true,"ca_institutions":"Université du Québec à Montréal","funders":"","keywords":"Machine translation; Computer science; Artificial intelligence; Natural language processing; Indigenous; Preprocessor; Context (archaeology); Focus (optics); Linguistics; History","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001038808,0.00007596976,0.0001220922,0.00002306221,0.00007100412,0.00004895057,0.0002810307,0.00003520126,0.000004250954],"category_scores_gemma":[0.00003822125,0.00004749445,0.00002179369,0.0003574092,0.00006480328,0.0005354454,0.00007920366,0.00007679642,8.556444e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000003763988,"about_ca_system_score_gemma":0.00002443509,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0002545674,"about_ca_topic_score_gemma":0.00003864026,"domain_scores_codex":[0.9994334,0.00004081417,0.0001968174,0.0001518634,0.0001010072,0.00007610011],"domain_scores_gemma":[0.99959,0.00004132533,0.0001280865,0.0001532525,0.00006648292,0.00002080284],"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":[0.00001677093,0.00002760504,0.0008225809,0.0006674188,0.00001315882,0.0001399935,0.04681382,0.00004984897,0.006437411,0.01155332,0.00004217257,0.9334159],"study_design_scores_gemma":[0.0008875453,0.0006085103,0.0005350981,0.0002931739,0.00009090674,0.002093927,0.001271369,0.7193475,0.2627408,0.01123165,0.0002785185,0.0006210536],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04172549,0.03815095,0.9176131,0.001790272,0.00001353646,0.0002180657,0.000004253286,0.0001917277,0.0002926233],"genre_scores_gemma":[0.824785,0.00002447321,0.1748225,0.0003394503,0.00001466621,0.000002092032,0.000001811223,0.000004034187,0.000006009639],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9327949,"threshold_uncertainty_score":0.1936767,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0221484901184939,"score_gpt":0.2994031450251255,"score_spread":0.2772546549066316,"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."}}