{"id":"W4407974294","doi":"10.1145/3720542","title":"A Hybrid Statistical and Rule-based Approach to Extremely Low-resource Machine Transliteration","year":2025,"lang":"en","type":"article","venue":"ACM Transactions on Asian and Low-Resource Language Information Processing","topic":"Algorithms and Data Compression","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Prince Edward Island","funders":"","keywords":"Transliteration; Computer science; Artificial intelligence; Rule-based system; Resource (disambiguation); Natural language processing; Machine learning; Data mining","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":[0.0002472473,0.0002428711,0.0002099243,0.0004652705,0.0005517892,0.0009099712,0.000417811,0.00008085258,0.00001693169],"category_scores_gemma":[0.00004368199,0.0002173727,0.00003932681,0.0004297072,0.00006257922,0.001422773,0.00003151516,0.0002812117,0.000008599008],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00003864958,"about_ca_system_score_gemma":0.00007625517,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001978154,"about_ca_topic_score_gemma":0.000003630306,"domain_scores_codex":[0.9985735,0.00007143783,0.000424,0.0003528463,0.0002995335,0.00027872],"domain_scores_gemma":[0.9991003,0.00008891781,0.00009443161,0.0004677429,0.00005766716,0.0001909157],"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":[0.00007260322,0.0001023241,0.000004399563,0.0002871466,0.00000973949,0.000003146618,0.005830551,0.001899835,0.0001315522,0.000458278,0.0001500535,0.9910504],"study_design_scores_gemma":[0.001732634,0.0001008589,0.0003989243,0.0005243962,0.00003528757,0.00004120888,0.001673437,0.9781325,0.002849063,0.0002993394,0.01376477,0.0004475872],"study_design_candidate":"design_other","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.003924785,0.0001910796,0.9906307,0.001369098,0.00003637658,0.0003349844,0.0001091963,0.000218749,0.003185048],"genre_scores_gemma":[0.841041,0.000004148892,0.1546967,0.003769673,0.00002337933,0.00006011467,0.00024004,0.00001298066,0.0001519452],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9906028,"threshold_uncertainty_score":0.8864201,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.005748411853017093,"score_gpt":0.2307737058908388,"score_spread":0.2250252940378217,"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."}}