{"id":"W7130558586","doi":"10.1109/fllm67465.2025.11390942","title":"Analysis of Linguistic Effects of Self-Consuming Training","year":2025,"lang":"","type":"article","venue":"","topic":"Mental Health via Writing","field":"Psychology","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Training (meteorology); Classifier (UML); Affect (linguistics); Training set; Language acquisition; Language model; Process (computing)","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":[0.006370684,0.0005950057,0.0004146365,0.0007257996,0.0005607602,0.0008604287,0.0008409967,0.0009166617,0.002210872],"category_scores_gemma":[0.05408822,0.0002858573,0.0004653364,0.000529482,0.001276641,0.001267881,0.001156856,0.001677564,0.0003897521],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008625371,"about_ca_system_score_gemma":0.0004370393,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003114694,"about_ca_topic_score_gemma":0.004094781,"domain_scores_codex":[0.9975551,0.001127309,0.0002060402,0.000426532,0.0004698533,0.000215131],"domain_scores_gemma":[0.9206902,0.06511053,0.002989744,0.006549767,0.003634002,0.001025866],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"observational","study_design_scores_codex":[0.003630177,0.002334151,0.14096,0.001326139,0.0006254538,0.001189105,0.002928516,0.5889697,0.06617104,0.005496805,0.007946648,0.1784223],"study_design_scores_gemma":[0.0001963257,0.002716351,0.1001372,0.0002408359,0.0003331186,0.0005983956,0.001378648,0.825488,0.05685161,0.007455649,0.004472504,0.0001313213],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.98555,0.000486033,0.01003089,0.0004521202,0.00008407751,0.00007332923,0.0003423418,0.0003940388,0.002587022],"genre_scores_gemma":[0.9930662,0.00009386016,0.004755706,0.0001614638,0.0000261917,0.00006461744,0.0006538665,0.0001217888,0.001056317],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.006370684,"threshold_uncertainty_score":0.03369176,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02896317986814428,"score_gpt":0.3943614385772697,"score_spread":0.3653982587091254,"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."}}