{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.00303479,0.001713422,0.001191225,0.003913297,0.001123289,0.001840074,0.001339043,0.001245334,0.002300305],"category_scores_gemma":[0.01301472,0.0004404939,0.001169646,0.002386089,0.0007339494,0.002493944,0.00134559,0.001582923,0.002836979],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008474285,"about_ca_system_score_gemma":0.001435844,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003590288,"about_ca_topic_score_gemma":0.008788537,"domain_scores_codex":[0.9965122,0.001042718,0.0002949238,0.001107397,0.0008834949,0.000159189],"domain_scores_gemma":[0.9948881,0.00140135,0.0004419214,0.0008755365,0.002226487,0.0001666672],"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":[0.0002630515,0.0001866582,0.002496505,0.0003815737,0.0002540033,0.0001018996,0.0004947621,0.02872385,0.0377805,0.00372694,0.0203875,0.9052027],"study_design_scores_gemma":[0.0001443518,0.0008705559,0.006433936,0.0001299573,0.0002466977,0.0005081691,0.0004076938,0.8733346,0.0614646,0.01485447,0.0414235,0.0001814667],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01395805,0.0009487821,0.9701161,0.0001787743,0.0001797752,0.000262541,0.0005318411,0.01202876,0.001795277],"genre_scores_gemma":[0.1711538,0.0003821513,0.8147351,0.0002507881,0.00029889,0.0007602969,0.004921751,0.001803108,0.00569417],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.003913297,"threshold_uncertainty_score":0.01604968,"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."}}