{"id":"W7162040362","doi":"10.36829/63cts.v12i2.1785","title":"Improving the Large Language Models (LMMs) Systems for Underrepresent Languages","year":2025,"lang":"","type":"article","venue":"Ciencia Tecnologí­a y Salud","topic":"Language and cultural evolution","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Scale AI","keywords":"Generative grammar; Indigenous; Knowledge base; Computational linguistics; Indigenous language; The Internet","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":[0.003390875,0.001387078,0.001463368,0.001112714,0.0009227822,0.00181295,0.00230574,0.001535327,0.004295605],"category_scores_gemma":[0.01482302,0.0005956551,0.001132418,0.0008839491,0.0007790973,0.003358203,0.002421177,0.003202095,0.002436995],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001105735,"about_ca_system_score_gemma":0.001652555,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00893282,"about_ca_topic_score_gemma":0.01447591,"domain_scores_codex":[0.9984731,0.0006286516,0.0001141654,0.000475161,0.0002172369,0.00009172364],"domain_scores_gemma":[0.9927851,0.005168908,0.0003676662,0.0007694805,0.0007901585,0.000118782],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.0004508968,0.0002732469,0.003322172,0.0003320631,0.0004112402,0.000284576,0.0008684134,0.2478172,0.007494073,0.01131836,0.006644382,0.7207834],"study_design_scores_gemma":[0.00002709452,0.00007746606,0.0004090983,0.00002374055,0.00007248111,0.00005903855,0.0001013616,0.9816005,0.002402764,0.01314356,0.002063583,0.00001927426],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.05008111,0.001691444,0.9368534,0.001019402,0.0002069321,0.0001384357,0.0003442598,0.006576225,0.003088704],"genre_scores_gemma":[0.546796,0.0006926851,0.4437915,0.0009126538,0.0002544321,0.0002908722,0.001206264,0.0005281098,0.005527528],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.00893282,"threshold_uncertainty_score":0.01793289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02246782002541666,"score_gpt":0.3160752349372846,"score_spread":0.293607414911868,"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."}}