{"id":"W4400526322","doi":"10.1145/3626772.3657952","title":"On Backbones and Training Regimes for Dense Retrieval in African Languages","year":2024,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Training (meteorology); Computer science; Natural language processing; Artificial intelligence; Geography; Meteorology","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":[0.007625329,0.00130517,0.001074546,0.001731155,0.001201528,0.001770259,0.001565427,0.00161588,0.004787194],"category_scores_gemma":[0.02193758,0.0006556642,0.0010401,0.001469721,0.001424604,0.006231873,0.003320444,0.002933566,0.004291962],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001059504,"about_ca_system_score_gemma":0.001471076,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008901303,"about_ca_topic_score_gemma":0.01216936,"domain_scores_codex":[0.9971468,0.001516294,0.0002381335,0.0005896165,0.0002608338,0.000248429],"domain_scores_gemma":[0.9913357,0.005627016,0.000330448,0.001860184,0.0006326654,0.0002138603],"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":[0.001987202,0.0008828722,0.01350673,0.001094041,0.0003389536,0.0004137554,0.001190321,0.1902858,0.02275395,0.01287582,0.03599589,0.7186747],"study_design_scores_gemma":[0.0003986415,0.0007647317,0.004776386,0.0003356812,0.0001980853,0.0005463024,0.0009084308,0.9228725,0.0256016,0.0279438,0.01555656,0.00009724161],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.5434727,0.01062036,0.3954862,0.003821022,0.0004347124,0.0006226195,0.005205717,0.01846059,0.02187611],"genre_scores_gemma":[0.8181439,0.002121192,0.1582602,0.0009230501,0.0002068352,0.0004612861,0.01356143,0.001158276,0.005163766],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.008901303,"threshold_uncertainty_score":0.04032707,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02033855501442258,"score_gpt":0.304658782273631,"score_spread":0.2843202272592085,"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."}}