{"id":"W7126374618","doi":"10.18653/v1/2024.law-1.10","title":"Enhancing Text Classification through LLM-Driven Active Learning and Human Annotation","year":2024,"lang":"","type":"article","venue":"","topic":"Machine Learning and Algorithms","field":"Computer Science","cited_by":1,"is_retracted":false,"has_abstract":true,"ca_institutions":"Ontario Tech University","funders":"","keywords":"Active learning (machine learning); Annotation; Focus (optics); Identification (biology)","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":[0.006268738,0.001816048,0.001622972,0.002114687,0.0009833552,0.002421736,0.004038853,0.002746258,0.002975157],"category_scores_gemma":[0.02012964,0.0005774106,0.001159173,0.001630043,0.001333182,0.005108928,0.003088419,0.002993462,0.002106007],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001187942,"about_ca_system_score_gemma":0.001164077,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002093203,"about_ca_topic_score_gemma":0.004329711,"domain_scores_codex":[0.9961074,0.00149637,0.000220802,0.0009764677,0.0009831882,0.0002157947],"domain_scores_gemma":[0.9849627,0.009908693,0.0008853046,0.001892745,0.002065125,0.0002854438],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.0006509101,0.0009197113,0.004063659,0.0003405071,0.0001141332,0.0001534919,0.0005096212,0.09360546,0.02394868,0.006028655,0.007629654,0.8620355],"study_design_scores_gemma":[0.00003032968,0.00008645784,0.0003971195,0.00002249276,0.00002345639,0.00005216405,0.00006430486,0.9782496,0.01135593,0.007815241,0.001884454,0.00001844357],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.01835627,0.0003151567,0.9746597,0.000415648,0.0001127182,0.0001631798,0.0001785025,0.004329341,0.001469431],"genre_scores_gemma":[0.4492967,0.0001938386,0.5422384,0.0007264097,0.0002712362,0.0006079912,0.00130832,0.0004815669,0.004875605],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.006268738,"threshold_uncertainty_score":0.03315264,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02140823191819688,"score_gpt":0.3204418519509792,"score_spread":0.2990336200327823,"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."}}