{"id":"W6929215141","doi":"10.48448/abca-vb49","title":"Large Language Models are biased to overestimate profoundness","year":2023,"lang":"en","type":"other","venue":"Open MIND","topic":"Blood properties and coagulation","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Natural (archaeology); Work (physics); Expression (computer science); Natural language","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":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.009246191,0.0008058858,0.000385829,0.0009407054,0.0002499393,0.001806083,0.0005568961,0.000540285,0.003148609],"category_scores_gemma":[0.08881871,0.0003028244,0.0004528314,0.0003482155,0.001240571,0.002607943,0.002045957,0.001170566,0.0006908729],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004980168,"about_ca_system_score_gemma":0.0003550698,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0008424974,"about_ca_topic_score_gemma":0.001298882,"domain_scores_codex":[0.9938357,0.003276266,0.0003575006,0.0009756569,0.001388308,0.0001665154],"domain_scores_gemma":[0.9193193,0.06096593,0.00879213,0.00617295,0.003588198,0.001161431],"domain_codex":null,"domain_gemma":"evaluation","domain_candidate":"evaluation","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"observational","study_design_scores_codex":[0.003610822,0.0004054454,0.3707482,0.001406983,0.0009696251,0.0007907483,0.01113617,0.03820824,0.0774211,0.02025587,0.00750813,0.4675386],"study_design_scores_gemma":[0.0002490848,0.00201619,0.4105925,0.0005432061,0.000580052,0.002430994,0.006450754,0.3408522,0.06213671,0.1492105,0.02449245,0.0004453907],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8836692,0.0007971937,0.09476612,0.000937613,0.000111079,0.0001408489,0.0004513865,0.001479924,0.01764672],"genre_scores_gemma":[0.988118,0.000122793,0.01019365,0.0002735187,0.00002545652,0.00003474685,0.00023368,0.00009499555,0.0009031753],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9907538,"threshold_uncertainty_score":0.04889911,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.08727980673928541,"score_gpt":0.3538016882916343,"score_spread":0.2665218815523489,"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."}}