{"id":"W4409299747","doi":"10.1016/j.mlwa.2025.100649","title":"Optimizing translation for low-resource languages: Efficient fine-tuning with custom prompt engineering in large language models","year":2025,"lang":"en","type":"article","venue":"Machine Learning with Applications","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":8,"is_retracted":false,"has_abstract":true,"ca_institutions":"Impact","funders":"University of Pretoria","keywords":"Computer science; Translation (biology); Resource (disambiguation); Chemistry","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003258263,0.0001933247,0.0001898367,0.0003448847,0.0002052605,0.0001350588,0.000487376,0.00006723326,0.000001430639],"category_scores_gemma":[0.00002973207,0.0001584354,0.00003289655,0.000997878,0.0000183756,0.0001882383,0.00007740958,0.0004256453,9.971694e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00006966244,"about_ca_system_score_gemma":0.00005745796,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004376139,"about_ca_topic_score_gemma":0.00007039937,"domain_scores_codex":[0.9988206,0.00003226926,0.0002049803,0.0004438679,0.0001687076,0.0003295747],"domain_scores_gemma":[0.9992608,0.0001693292,0.00009893945,0.0003648634,0.00006141019,0.00004465501],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00004223212,0.0001082481,0.0003850463,0.0002546478,0.00002239062,0.000007211907,0.004585239,0.9158376,0.003106774,0.04204055,0.000009928584,0.03360007],"study_design_scores_gemma":[0.0008165436,0.00005075882,0.00002636089,0.0002928885,0.00001387198,0.000005502304,0.0001260695,0.9946553,0.002716336,0.0001736077,0.0009146263,0.0002081987],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.004841454,0.001715146,0.9908051,0.0004068984,0.000004802539,0.0009539213,0.000004955128,0.0008701376,0.0003975573],"genre_scores_gemma":[0.535341,9.668198e-7,0.4638509,0.00005144806,0.0000125446,0.0005773879,0.00004116261,0.00001864132,0.0001058551],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5304996,"threshold_uncertainty_score":0.646081,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005241564749280432,"score_gpt":0.2498799504021441,"score_spread":0.2446383856528637,"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."}}