{"id":"W4412889455","doi":"10.18653/v1/2025.acl-srw.5","title":"Rethinking Full Finetuning from Pretraining Checkpoints in Active Learning for African Languages","year":2025,"lang":"en","type":"article","venue":"","topic":"Intelligent Tutoring Systems and Adaptive Learning","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research; Nvidia","keywords":"Computer science; Natural language processing; Artificial intelligence; Programming language","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":[0.0006242723,0.0001608499,0.0002438973,0.0002498475,0.0001961327,0.0001843158,0.0005086655,0.0000921142,0.0000203761],"category_scores_gemma":[0.0005394392,0.0001551108,0.00007805545,0.0004201655,0.00001712156,0.0003279745,0.0002017948,0.0004339556,0.000008563454],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001234748,"about_ca_system_score_gemma":0.00007216771,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0007149795,"about_ca_topic_score_gemma":0.0001463377,"domain_scores_codex":[0.9985636,0.0001252062,0.0002971725,0.0004819874,0.0001631629,0.0003688664],"domain_scores_gemma":[0.9988225,0.0007012288,0.0001245548,0.0002346251,0.0000794012,0.00003766807],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.00009916336,0.00006454634,0.009954197,0.00009944788,0.0001585365,0.00004555213,0.08497003,0.01164166,0.01923854,0.5751672,0.0001936974,0.2983674],"study_design_scores_gemma":[0.002228932,0.0004240372,0.0205972,0.002882115,0.00002926028,0.000006563397,0.02846465,0.8192202,0.04048398,0.02657443,0.05779779,0.001290852],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.07676202,0.0001392471,0.889212,0.0004395391,0.0003493107,0.0002483992,0.000001002806,0.0002639103,0.03258459],"genre_scores_gemma":[0.9311847,0.000002234801,0.05678455,0.0001183837,0.000100825,0.00003570377,0.000004876355,0.00001138166,0.01175732],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8544227,"threshold_uncertainty_score":0.6325233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02508123280539928,"score_gpt":0.2826568588096516,"score_spread":0.2575756260042523,"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."}}