{"id":"W4229014795","doi":"10.18653/v1/2022.naacl-main.223","title":"A Few Thousand Translations Go a Long Way! Leveraging Pre-trained Models for African News Translation","year":2022,"lang":"en","type":"article","venue":"Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":45,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"International Development Research Centre; Rockefeller Foundation","keywords":"Blessing; Art history; Art; Theology; Philosophy; Classics","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":[0.0005129569,0.0002058844,0.0003764364,0.000195959,0.0006522875,0.00007096353,0.002702472,0.00005276027,0.000002151499],"category_scores_gemma":[0.001283658,0.000144814,0.0003355903,0.0007955944,0.0003058274,0.0001372857,0.0005975296,0.0003153804,4.04962e-8],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001577812,"about_ca_system_score_gemma":0.0001087299,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001161386,"about_ca_topic_score_gemma":0.00009183045,"domain_scores_codex":[0.9980645,0.00003353878,0.0005922518,0.0003351099,0.0007245766,0.0002500826],"domain_scores_gemma":[0.9960173,0.0003700006,0.002103662,0.0003218968,0.001169382,0.00001772153],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0002089286,0.0003481233,0.02498404,0.0006803671,0.0005454106,3.062344e-7,0.03207278,0.04172688,0.01060368,0.8322361,0.0005182481,0.0560752],"study_design_scores_gemma":[0.001174129,0.0004364545,0.008136427,0.0002168069,0.0003118845,0.000005226057,0.00404992,0.2470319,0.02161998,0.716061,0.0003574356,0.0005988833],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4884732,0.001350877,0.4841038,0.01361231,0.0008665649,0.006691811,0.001571499,0.001502855,0.001827089],"genre_scores_gemma":[0.9291441,0.000003258096,0.0704291,0.00005510609,0.00003178962,0.0001635269,0.00001189636,0.00001964412,0.0001415522],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.4406709,"threshold_uncertainty_score":0.5905344,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02649600142926935,"score_gpt":0.2694050153342648,"score_spread":0.2429090139049955,"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."}}