{"id":"W4406735064","doi":"10.18653/v1/2025.emnlp-main.1413","title":"AFRIDOC-MT: Document-level MT Corpus for African Languages","year":2025,"lang":"en","type":"article","venue":"","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University; Canadian Institute for Advanced Research","funders":"Agence Nationale de la Recherche; International Development Research Centre; Rockefeller Foundation","keywords":"Linguistics; Natural language processing; Computer science; History; Philosophy","routes":{"ca_aff":true,"ca_fund":true,"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":[0.001802064,0.001114619,0.0007198725,0.004531565,0.001847751,0.001586931,0.001493535,0.001391057,0.02295722],"category_scores_gemma":[0.009171958,0.0005797604,0.0003352498,0.004135392,0.000582254,0.003283504,0.002744651,0.001626434,0.01291533],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0006286767,"about_ca_system_score_gemma":0.002522667,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.005855454,"about_ca_topic_score_gemma":0.006708491,"domain_scores_codex":[0.9990069,0.000289865,0.0001697431,0.0002173392,0.0002183743,0.00009777688],"domain_scores_gemma":[0.9961332,0.001403591,0.0003138882,0.0007547064,0.001065055,0.0003296046],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.001538114,0.0003214309,0.003268547,0.004509896,0.00007904164,0.001216653,0.001691547,0.002664693,0.02899292,0.01585965,0.5896085,0.350249],"study_design_scores_gemma":[0.0005752713,0.0001621108,0.01115617,0.0005355433,0.000100841,0.001349821,0.0008204142,0.008585989,0.02731806,0.009254573,0.9400187,0.0001225508],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"dataset","genre_gemma":"empirical","genre_scores_codex":[0.08262798,0.006285604,0.08057711,0.003527172,0.00179131,0.001554211,0.7210131,0.04562022,0.0570033],"genre_scores_gemma":[0.1159195,0.00138339,0.08557712,0.0003308016,0.0002439887,0.001418821,0.7794691,0.005397561,0.01025956],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.02295722,"threshold_uncertainty_score":0.07679951,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01478806632869473,"score_gpt":0.3071332285489138,"score_spread":0.2923451622202191,"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."}}