{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.004118209,0.002746998,0.002078208,0.002087673,0.00150701,0.004249615,0.002035005,0.002703636,0.01445381],"category_scores_gemma":[0.01550593,0.001622388,0.001758924,0.002749872,0.0009019708,0.009641599,0.002746585,0.004339157,0.02848225],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009443113,"about_ca_system_score_gemma":0.001272563,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00679284,"about_ca_topic_score_gemma":0.01176919,"domain_scores_codex":[0.9977717,0.0009298298,0.0001710908,0.000592505,0.0003621736,0.0001727867],"domain_scores_gemma":[0.9934442,0.00343449,0.0001946177,0.001606359,0.001101727,0.0002187],"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":[0.001190602,0.0002715682,0.00258473,0.0005194543,0.0005409976,0.0005675459,0.0003813801,0.05199802,0.006544474,0.004243212,0.1742343,0.7569237],"study_design_scores_gemma":[0.0002513161,0.0004100902,0.002091589,0.0003946863,0.0004659991,0.000664329,0.0006932465,0.8371603,0.01242836,0.04419842,0.101076,0.0001657231],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1223586,0.04524403,0.6476969,0.02604831,0.01649205,0.0004380156,0.01289876,0.08603851,0.04278486],"genre_scores_gemma":[0.462193,0.01227112,0.3818915,0.006071209,0.005303365,0.0004748489,0.05432587,0.008704035,0.06876514],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.01445381,"threshold_uncertainty_score":0.0483529,"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."}}