{"id":"W4400374397","doi":"10.48550/arxiv.2407.02320","title":"Exploring the Role of Transliteration in In-Context Learning for Low-resource Languages Written in Non-Latin Scripts","year":2024,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Second Language Learning and Teaching","field":"Arts and Humanities","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Atomic Energy of Canada Limited; Strong; Institute for Catastrophic Loss Reduction","keywords":"Transliteration; Scripting language; Context (archaeology); Computer science; Natural language processing; Resource (disambiguation); Artificial intelligence; Linguistics; History; Programming language","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":[0.002064456,0.001513073,0.0007331563,0.000501739,0.0007109614,0.001471154,0.001171332,0.00113358,0.002880922],"category_scores_gemma":[0.009346644,0.0003732366,0.0007318189,0.0006249421,0.0007383075,0.003454879,0.001375239,0.002185321,0.001797837],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008070048,"about_ca_system_score_gemma":0.001414706,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004445002,"about_ca_topic_score_gemma":0.008802913,"domain_scores_codex":[0.9988133,0.0006250722,0.00005270477,0.0003741694,0.00006160414,0.00007316168],"domain_scores_gemma":[0.9950381,0.003601759,0.0002005329,0.0006382567,0.0003257216,0.0001955598],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0019172,0.0007491575,0.01263643,0.0009591965,0.0002517405,0.0007922992,0.001725826,0.2099239,0.04040401,0.009618544,0.01421046,0.7068111],"study_design_scores_gemma":[0.00008327823,0.0002964139,0.0009698879,0.00004503677,0.00006644619,0.0001695925,0.0003022612,0.9624803,0.02363469,0.008385888,0.003531313,0.00003478161],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4439415,0.002582049,0.5269724,0.001381005,0.0003413949,0.0002737292,0.0008881856,0.01536763,0.008252035],"genre_scores_gemma":[0.8405077,0.0003604594,0.1520222,0.000490614,0.00007658143,0.0001903824,0.001750769,0.0007717106,0.003829486],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.004445002,"threshold_uncertainty_score":0.01091802,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0813258279862046,"score_gpt":0.1920477594658318,"score_spread":0.1107219314796272,"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."}}