{"id":"W7106855247","doi":"10.48448/fqkf-dv48","title":"AfroXLMR-Social: Adapting Pre-trained Language Models for African Languages Social Media Text","year":2025,"lang":"","type":"other","venue":"Open MIND","topic":"","field":"","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Social media; Task (project management); Baseline (sea); Language model; Domain (mathematical analysis); Diversity (politics); Training set","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":[0.002268263,0.002444796,0.0006750246,0.001557981,0.0009426245,0.001341654,0.001840185,0.001544388,0.008537572],"category_scores_gemma":[0.006991966,0.0007023511,0.001165024,0.0009308991,0.0006014658,0.002959433,0.002671935,0.0033255,0.01035719],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.000865783,"about_ca_system_score_gemma":0.001437724,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01215294,"about_ca_topic_score_gemma":0.01918805,"domain_scores_codex":[0.9986355,0.0005789273,0.00007791797,0.0004030307,0.0001775961,0.0001269762],"domain_scores_gemma":[0.9977213,0.001076271,0.00009069516,0.0005187012,0.0004903121,0.0001027721],"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":[0.000738978,0.0008319218,0.007288692,0.0006826138,0.0003504097,0.0007586742,0.001015424,0.05158939,0.02842025,0.002855157,0.1193445,0.7861241],"study_design_scores_gemma":[0.0002243523,0.0004111074,0.006019555,0.0001964689,0.0001362635,0.0006238662,0.001098075,0.8718705,0.04156466,0.005503949,0.07216962,0.000181536],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.316722,0.004599695,0.4915617,0.003132214,0.003597372,0.001102217,0.02699505,0.1275075,0.02478229],"genre_scores_gemma":[0.5516906,0.001239042,0.337225,0.001369095,0.0004747697,0.00129081,0.08067077,0.004609682,0.02143026],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01215294,"threshold_uncertainty_score":0.028561,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05104030918918349,"score_gpt":0.3548779572158284,"score_spread":0.3038376480266449,"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."}}