{"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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001379459,0.0003040823,0.0003963067,0.0001725158,0.0001780992,0.00008152496,0.001820731,0.0000920902,0.0002540679],"category_scores_gemma":[0.001413344,0.0002147809,0.00008367409,0.000335949,0.0008287659,0.0001521111,0.0007995956,0.0002840704,0.000009575704],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004643306,"about_ca_system_score_gemma":0.0002999025,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.001097439,"about_ca_topic_score_gemma":0.0003344427,"domain_scores_codex":[0.9976728,0.0001005189,0.000392324,0.0009349771,0.0004519885,0.0004474128],"domain_scores_gemma":[0.9975154,0.000527537,0.0003517449,0.001300126,0.0002181386,0.00008699125],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"bench_or_experimental","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001812034,0.004179126,0.001120894,0.01665391,0.001373517,0.0006306767,0.07468782,0.004306952,0.6219754,0.008399137,0.02312185,0.2417387],"study_design_scores_gemma":[0.002852674,0.002351712,0.0003660401,0.005221565,0.0002765296,0.00009089453,0.02335702,0.9328585,0.009722834,0.001295594,0.01926477,0.002341919],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5414882,0.003244762,0.02702745,0.0002719505,0.001368009,0.005126942,0.003672396,0.001373123,0.4164271],"genre_scores_gemma":[0.7241846,0.0000335068,0.1712352,0.0001113667,0.001187961,0.000169197,0.0005715827,0.0003958531,0.1021107],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9285515,"threshold_uncertainty_score":0.8758513,"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."}}