{"id":"W4406883348","doi":"10.1142/s2717554524500176","title":"The Power of Personalized Datasets: Advancing Chinese Composition Writing for Elementary School through Targeted Model Fine-Tuning","year":2025,"lang":"en","type":"article","venue":"International Journal of Asian Language Processing","topic":"Educational Technology and Assessment","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Ottawa","funders":"","keywords":"Composition (language); Power (physics); Mathematics education; Computer science; Psychology; Physics; Literature; Art","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":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0005214539,0.00009718375,0.000137239,0.0001576315,0.0002043033,0.0001337484,0.0009533321,0.00003595629,0.000007199657],"category_scores_gemma":[0.0002174028,0.00007232232,0.00008094208,0.0001891264,0.00005792538,0.0009318,0.000126889,0.0002037675,2.552032e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001207536,"about_ca_system_score_gemma":0.0003864524,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000003163849,"about_ca_topic_score_gemma":0.000003295441,"domain_scores_codex":[0.9988838,0.00003178269,0.0004842594,0.0001320163,0.0003263372,0.0001418201],"domain_scores_gemma":[0.998701,0.0001588792,0.0005109192,0.0001266283,0.0004749106,0.00002771355],"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":[0.0008587337,0.001193243,0.01616094,0.0003939758,0.001708626,0.000139736,0.02412187,0.008638265,0.219713,0.1370912,0.009171224,0.5808092],"study_design_scores_gemma":[0.009763278,0.0004550471,0.008402437,0.006847167,0.0002193327,0.0006607915,0.05047804,0.6890465,0.05761943,0.1720396,0.003406994,0.001061419],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.06308648,0.002818387,0.9232937,0.009979052,0.00035315,0.00009848734,0.00003449625,0.00002133397,0.0003149896],"genre_scores_gemma":[0.7428207,0.00001374944,0.2566509,0.0003856668,0.00006943546,0.000006327944,0.00002647961,0.000004195717,0.00002260163],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6804082,"threshold_uncertainty_score":0.2949219,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.007021506712353659,"score_gpt":0.3488968130148321,"score_spread":0.3418753063024784,"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."}}