{"id":"W6929420084","doi":"10.48448/000a-es66","title":"Small Data? No Problem: Exploring the Viability of Multilingual Pretrained Language Models for Low-resourced Languages","year":2021,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Animal testing and alternatives","field":"Veterinary","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Language model; Joint (building); Variety (cybernetics); Training set; Constructed language; Multilingualism; Work (physics)","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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.006083204,0.002184783,0.001111819,0.001067796,0.00112857,0.002808429,0.003301926,0.001541859,0.006317537],"category_scores_gemma":[0.0175558,0.0008909128,0.00171186,0.001243846,0.001590457,0.008899432,0.004379365,0.004139408,0.004039923],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001121992,"about_ca_system_score_gemma":0.001691843,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0147409,"about_ca_topic_score_gemma":0.02395388,"domain_scores_codex":[0.9973434,0.001310637,0.0001371597,0.0007842471,0.0002293998,0.0001952807],"domain_scores_gemma":[0.9911422,0.005931191,0.0001951545,0.001657369,0.0006978048,0.0003762519],"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":[0.003034453,0.001029875,0.02265391,0.001143091,0.001130438,0.001184566,0.001575896,0.3388433,0.01188316,0.02539065,0.09575126,0.4963794],"study_design_scores_gemma":[0.0002477521,0.0002932711,0.001451293,0.0001256193,0.0001383648,0.0001941297,0.000608101,0.9490261,0.00668179,0.02711528,0.01404538,0.00007298339],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.4089683,0.005880436,0.5028034,0.01241437,0.001520299,0.0004966984,0.01020954,0.02634362,0.03136321],"genre_scores_gemma":[0.7431738,0.001121534,0.215482,0.002839538,0.0003441657,0.0005251386,0.02412139,0.003107816,0.009284667],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0147409,"threshold_uncertainty_score":0.03217143,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.3492445313822033,"score_gpt":0.4170720861778187,"score_spread":0.06782755479561542,"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."}}