{"id":"W7077856092","doi":"10.48448/d6cr-4k73","title":"A Guide To Effectively Leveraging LLMs for Low-Resource Text Summarization: Data Augmentation and Semi-supervised Approaches","year":2025,"lang":"en","type":"other","venue":"Underline Science Inc.","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo; McGill University","funders":"","keywords":"Automatic summarization; Inference; Quality (philosophy); Measure (data warehouse); Language model; Readability; Labeled data; Codebase","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":[0.002308835,0.002590665,0.000975713,0.002746815,0.0007693688,0.002239089,0.002604476,0.001960864,0.008661934],"category_scores_gemma":[0.01010498,0.001300593,0.001245348,0.002379512,0.0008904698,0.00320134,0.002393362,0.00501004,0.0123637],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009333294,"about_ca_system_score_gemma":0.00146373,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003034351,"about_ca_topic_score_gemma":0.007580351,"domain_scores_codex":[0.9982168,0.0006796547,0.0001904517,0.0004895165,0.0003605964,0.00006301518],"domain_scores_gemma":[0.9958281,0.002005089,0.0002753781,0.000692039,0.001032051,0.000167231],"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":[0.0002128608,0.0002481978,0.0007856937,0.001386573,0.0001529273,0.0003182933,0.0006588449,0.03687244,0.02443347,0.02075703,0.1149656,0.799208],"study_design_scores_gemma":[0.0001674628,0.0002998362,0.0008688499,0.0004482256,0.00006954562,0.0005065661,0.0003047876,0.6403252,0.03891259,0.07337847,0.2445022,0.0002162054],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.000962498,0.001545692,0.9787865,0.0006424375,0.0001201266,0.0002168581,0.001309632,0.01517561,0.001240675],"genre_scores_gemma":[0.009000756,0.001098223,0.9820642,0.0003942928,0.0001082277,0.0006273325,0.003173359,0.001093525,0.002440079],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.008661934,"threshold_uncertainty_score":0.02897704,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0479054683704552,"score_gpt":0.289259888969461,"score_spread":0.2413544205990058,"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."}}