{"id":"W4393228430","doi":"10.1002/ail2.92","title":"Building Text and Speech Benchmark Datasets and Models for Low‐Resourced East African Languages: Experiences and Lessons","year":2024,"lang":"en","type":"article","venue":"Applied AI Letters","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":9,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"FP7 International Cooperation; Deutsche Gesellschaft für Internationale Zusammenarbeit; International Development Research Centre; Rockefeller Foundation","keywords":"Benchmark (surveying); Computer science; Linguistics; Natural language processing; Artificial intelligence; Geography; Cartography","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0002030588,0.0001797405,0.0001636821,0.0001105107,0.0001591183,0.000688453,0.0003657039,0.00005597847,0.000001797174],"category_scores_gemma":[0.00001301989,0.0001513025,0.00001807686,0.0002006893,0.0001597976,0.0004699127,0.0003473907,0.0001543212,5.926997e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001753867,"about_ca_system_score_gemma":0.00001445001,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003025208,"about_ca_topic_score_gemma":0.000003707093,"domain_scores_codex":[0.998814,0.0000135694,0.00013617,0.0006068792,0.0001567589,0.0002725809],"domain_scores_gemma":[0.9994757,0.0001347167,0.0000376487,0.0002537949,0.000009443646,0.00008870052],"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":[0.00002968578,0.00002517368,0.00001227582,0.0005455198,0.00005617026,0.00013215,0.03035337,0.00002327407,0.2015778,0.2536163,0.01881117,0.4948171],"study_design_scores_gemma":[0.002276963,0.0003067032,0.0001076699,0.001630348,0.0001908405,0.0008393999,0.01576799,0.566207,0.1796246,0.1888831,0.0399924,0.004173049],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1839266,0.00609872,0.7967276,0.01192999,0.00006976359,0.0004684439,0.00006724805,0.0005796093,0.0001320249],"genre_scores_gemma":[0.7094385,0.00001271028,0.288442,0.001945399,0.00004227903,0.00008845529,0.0000130447,0.00001200692,0.000005628859],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.5661837,"threshold_uncertainty_score":0.6638768,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01178532502583053,"score_gpt":0.2778462994356153,"score_spread":0.2660609744097848,"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."}}