{"id":"W4386724574","doi":"10.31234/osf.io/x2f4a","title":"Shadows of wisdom: Classifying meta-cognitive and morally-grounded narrative content via Large Language Models","year":2023,"lang":"en","type":"preprint","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Narrative; Humility; Psychology; Categorization; Classifier (UML); Workflow; Social psychology; Computer science; Artificial intelligence; Linguistics; Political science","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":true,"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":[0.007049048,0.0009632317,0.0003812576,0.00277591,0.0009126024,0.003000067,0.001398831,0.0008632639,0.001676788],"category_scores_gemma":[0.03456535,0.0003600329,0.001146975,0.001082381,0.001015867,0.003575774,0.002304216,0.001821299,0.0007944094],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002215663,"about_ca_system_score_gemma":0.002015256,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01410669,"about_ca_topic_score_gemma":0.02386384,"domain_scores_codex":[0.997178,0.001784571,0.0001473521,0.0004635289,0.0003157028,0.0001107837],"domain_scores_gemma":[0.9706654,0.02426546,0.001318316,0.00173196,0.001548372,0.0004704759],"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":[0.001768333,0.0007060638,0.1700503,0.001849353,0.0005818772,0.0009045035,0.06834864,0.07482251,0.02227889,0.02851403,0.01523435,0.6149411],"study_design_scores_gemma":[0.00007530327,0.0001882426,0.03210264,0.0003028603,0.0001805362,0.0002739839,0.01025435,0.8895992,0.01195938,0.0395393,0.01537299,0.0001512233],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6674873,0.0008247241,0.3160304,0.001394593,0.00010118,0.0007049253,0.004486202,0.003507773,0.005462847],"genre_scores_gemma":[0.8524138,0.0001426386,0.1419127,0.0001138835,0.0000280678,0.0003503759,0.003784472,0.0001900388,0.001063989],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01410669,"threshold_uncertainty_score":0.03727943,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.231357794968577,"score_gpt":0.3276528238762003,"score_spread":0.09629502890762323,"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."}}