{"id":"W6891660818","doi":"10.48448/0r09-t174","title":"Predicting Machine Translation Performance on Low-Resource Languages: The Role of Domain Similarity","year":2024,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Machine translation; Similarity (geometry); Domain (mathematical analysis); Focus (optics); Translation (biology); Natural language; Language model","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"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":[0.003334696,0.0009183786,0.0006072242,0.00159399,0.0006152517,0.002133108,0.0006896168,0.0007782836,0.004318623],"category_scores_gemma":[0.02107227,0.0002018669,0.0004528452,0.002085733,0.0006499586,0.002202972,0.001389722,0.000913431,0.004027307],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006609512,"about_ca_system_score_gemma":0.0008488359,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.007135046,"about_ca_topic_score_gemma":0.007809608,"domain_scores_codex":[0.9980817,0.000985533,0.00009030959,0.0004653639,0.0002582624,0.0001187866],"domain_scores_gemma":[0.9891123,0.007341826,0.0008121338,0.001526165,0.0008751334,0.0003323542],"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":[0.001106374,0.0005934603,0.2186104,0.0006048832,0.0003914611,0.000810503,0.0008161613,0.2453182,0.02046121,0.004230288,0.02088317,0.4861739],"study_design_scores_gemma":[0.00009073828,0.0002507795,0.1043063,0.0001501418,0.0001148821,0.000464948,0.000971279,0.8527521,0.02046323,0.01202369,0.00830324,0.0001086476],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9196422,0.002086284,0.0564131,0.001248888,0.0001103699,0.00008244977,0.002952929,0.002317557,0.01514622],"genre_scores_gemma":[0.9748674,0.0002812853,0.01877834,0.0001278492,0.00004572489,0.0000373582,0.003661066,0.0003405421,0.001860542],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.007135046,"threshold_uncertainty_score":0.01763582,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0122021436285633,"score_gpt":0.2735886007659061,"score_spread":0.2613864571373428,"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."}}