{"id":"W3213418658","doi":"10.18653/v1/2021.mrl-1.11","title":"Small Data? No Problem! Exploring the Viability of Pretrained Multilingual Language Models for Low-resourced Languages","year":2021,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":114,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Natural Sciences and Engineering Research Council of Canada; Compute Canada","keywords":"Computer science; Natural language processing; Artificial intelligence; Language model; Second-generation programming language; Code (set theory); Resource (disambiguation); Variety (cybernetics); Training set; Programming language; Set (abstract data type); Fifth-generation programming language; Programming paradigm","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007922602,0.0001321014,0.0001992884,0.00003081077,0.00007931866,0.0000791637,0.001504796,0.00004271862,0.00001588119],"category_scores_gemma":[0.0004397654,0.00009428202,0.0000751871,0.0001632434,0.00003599158,0.0004000146,0.0009214114,0.0001187072,0.000002946975],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001894556,"about_ca_system_score_gemma":0.0001151617,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001795912,"about_ca_topic_score_gemma":0.0001804418,"domain_scores_codex":[0.9984944,0.0000873208,0.000339022,0.0005833096,0.000217746,0.0002781776],"domain_scores_gemma":[0.9972463,0.0004192292,0.00007997225,0.001991112,0.000208439,0.00005491379],"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":[0.0001225844,0.0009678908,0.0005877927,0.002209976,0.0003337996,0.00008172863,0.116352,0.1588636,0.1346414,0.1103087,0.0004585504,0.4750721],"study_design_scores_gemma":[0.0003590313,0.00002055823,0.00004200616,0.00003325785,0.000008424973,0.000003171357,0.001169819,0.9640996,0.03301531,0.0009735377,0.0001505,0.000124708],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3037039,0.00009988488,0.693785,0.0002707948,0.0001111462,0.0003659025,0.00002320122,0.0001264294,0.0015138],"genre_scores_gemma":[0.6142579,0.000002363363,0.3848851,0.0001229119,0.00007986397,0.00004859194,0.00001866829,0.00001015274,0.00057441],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8052361,"threshold_uncertainty_score":0.3844709,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1216007999799113,"score_gpt":0.2963027386983271,"score_spread":0.1747019387184158,"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."}}