{"id":"W1551794154","doi":"","title":"Transliteration Generation and Mining with Limited Training Resources","year":2010,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":39,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Alberta","funders":"","keywords":"Transliteration; Computer science; Discriminative model; Artificial intelligence; Natural language processing; Joint (building); Training set; Sequence (biology); n-gram; Language model; Machine learning; Engineering","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":[0.000153808,0.00006717636,0.00005658462,0.0000715896,0.000101095,0.0003513344,0.0001526386,0.00004603544,0.000004720825],"category_scores_gemma":[0.00001842942,0.00004734567,0.00000701377,0.0001511377,0.00002563079,0.000605419,0.00001906205,0.000108337,3.173573e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00000266767,"about_ca_system_score_gemma":0.00001391639,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000007125681,"about_ca_topic_score_gemma":0.00009733517,"domain_scores_codex":[0.9995077,0.00001504241,0.00008093548,0.0001962512,0.0001053504,0.00009476369],"domain_scores_gemma":[0.9997251,0.00002511189,0.00003051991,0.0001402937,0.00004464198,0.00003435384],"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":[0.000005444011,0.00001038371,0.0003429748,0.000009278471,0.000005214545,0.00000851577,0.02224577,0.00000351525,0.4253463,0.04545482,0.00009548053,0.5064722],"study_design_scores_gemma":[0.0008417395,0.0004449752,0.001197392,0.00008809853,0.00001732082,0.0003773462,0.0004244408,0.6354696,0.3497557,0.006420685,0.004184206,0.0007784882],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4741784,0.0001004523,0.5242534,0.0006100665,0.00003086242,0.00004617833,1.35677e-7,0.0003329594,0.0004474928],"genre_scores_gemma":[0.5123429,5.876227e-7,0.4874114,0.0001494768,0.00003771739,0.000003357178,0.000001218291,0.000002535651,0.0000508101],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6354661,"threshold_uncertainty_score":0.3387926,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01945453074298198,"score_gpt":0.2420105028488689,"score_spread":0.2225559721058869,"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."}}