{"id":"W4410595593","doi":"10.1016/j.mlwa.2025.100666","title":"A novel unsupervised fine-tuning method for text summarization, and highlighting the limitations of ROUGE score","year":2025,"lang":"en","type":"article","venue":"Machine Learning with Applications","topic":"Topic Modeling","field":"Computer Science","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Windsor","funders":"","keywords":"Automatic summarization; ROUGE; Artificial intelligence; Computer science; Natural language processing; Machine learning","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":[0.0003996001,0.0001062913,0.0001352171,0.000120194,0.0005636725,0.00009296591,0.0004196296,0.00003541642,8.57971e-7],"category_scores_gemma":[0.0002393471,0.00007804261,0.00002961537,0.0007120727,0.00004035618,0.0001230197,0.0001389304,0.0001675788,5.741921e-7],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00001501493,"about_ca_system_score_gemma":0.00008006213,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001067094,"about_ca_topic_score_gemma":0.00009613877,"domain_scores_codex":[0.9991354,0.00005281791,0.0002420695,0.0003167248,0.000108159,0.0001448829],"domain_scores_gemma":[0.9980642,0.001169709,0.0001413025,0.0004022853,0.0001922404,0.00003030577],"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":[0.000005925544,0.00004058434,0.007132932,0.00005196332,0.00004025811,5.577025e-8,0.00118862,0.056687,0.004212661,0.7755992,0.00002329431,0.1550175],"study_design_scores_gemma":[0.0004138095,0.00002686135,0.002615154,0.00004167061,0.0000336449,0.00000285859,0.00007628017,0.9743517,0.0002877192,0.002576283,0.01948162,0.00009237471],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.0009535243,0.0001859692,0.990668,0.006722608,0.00001715368,0.0006108287,0.000004790667,0.0001086469,0.0007285057],"genre_scores_gemma":[0.3034286,0.00001262459,0.6954219,0.0001718579,0.00002098648,0.0004387152,0.00002106349,0.000009471778,0.000474832],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.9176647,"threshold_uncertainty_score":0.433537,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03293311905183436,"score_gpt":0.2762938464379979,"score_spread":0.2433607273861635,"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."}}