{"id":"W4378418331","doi":"10.18280/ria.370219","title":"BalBERT: A New Approach to Improving Dataset Balancing for Text Classification","year":2023,"lang":"en","type":"article","venue":"Revue d intelligence artificielle","topic":"Text and Document Classification Technologies","field":"Computer Science","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"","keywords":"Computer science; Natural language processing; Information retrieval; Artificial intelligence","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.000562489,0.0001690814,0.000177945,0.0003266388,0.0002674556,0.0003160276,0.001500559,0.00009647251,0.00002364654],"category_scores_gemma":[0.000530634,0.0001683846,0.00007186236,0.001659898,0.00003973786,0.0004806919,0.0003569083,0.0001182638,0.002432049],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00007310254,"about_ca_system_score_gemma":0.00008101042,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00004240864,"about_ca_topic_score_gemma":0.000006326123,"domain_scores_codex":[0.9981253,0.00002362802,0.0004242863,0.0007835951,0.0001872099,0.000455958],"domain_scores_gemma":[0.9980976,0.0002497865,0.0001329499,0.00129592,0.00007780301,0.0001459132],"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":[0.000009453669,0.00008078614,0.00006232543,0.0000711818,0.000009831642,0.000001029061,0.001203133,0.005045435,0.02172928,0.2524338,0.1580365,0.5613173],"study_design_scores_gemma":[0.00004320373,0.00005075712,0.0001368452,0.00002019234,0.000005221778,0.000004415799,0.0008694687,0.8364382,0.02721041,0.006376718,0.1286027,0.0002418169],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.0006009972,0.00003264205,0.9896689,0.005868196,0.000297297,0.0006591339,0.00003881934,0.0008113375,0.002022663],"genre_scores_gemma":[0.7852168,0.00005069737,0.1955422,0.0007213714,0.0002391522,0.0006276763,0.0006701998,0.00004632141,0.01688555],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8313928,"threshold_uncertainty_score":0.9983447,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09182440662175224,"score_gpt":0.3138121450268641,"score_spread":0.2219877384051118,"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."}}