{"id":"W7080350807","doi":"","title":"One-size-fits-all Evaluation of LLMs for Safety Assurance Considered Harmful","year":2025,"lang":"en","type":"article","venue":"HAL (Le Centre pour la Communication Scientifique Directe)","topic":"Geochemistry and Geologic Mapping","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Canada Research Chairs; University of Toronto; Critical Systems Labs","funders":"","keywords":"Risk assessment; Risk management; Government (linguistics); Public health; MEDLINE","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":[0.00567381,0.00179708,0.001544065,0.001510693,0.0006720271,0.001873532,0.001961211,0.001339896,0.01197782],"category_scores_gemma":[0.01887807,0.0002326267,0.001799224,0.000684096,0.0009697542,0.002390564,0.001373107,0.000900014,0.0007963759],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002550597,"about_ca_system_score_gemma":0.002106919,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004888847,"about_ca_topic_score_gemma":0.005450691,"domain_scores_codex":[0.9960667,0.001671396,0.000119563,0.0004316598,0.001199713,0.0005109248],"domain_scores_gemma":[0.9799737,0.0137318,0.0008466793,0.001784722,0.002959614,0.0007034863],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.005784572,0.0005528646,0.007047159,0.001000133,0.0003668886,0.0002632935,0.0001313905,0.677519,0.008507855,0.01282247,0.006345332,0.2796591],"study_design_scores_gemma":[0.0001290511,0.003192433,0.003114846,0.00009500059,0.0002884374,0.0002125631,0.0003755473,0.96547,0.009067696,0.01361257,0.004402798,0.00003911848],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6342993,0.003727914,0.2982295,0.001750571,0.0005088993,0.000787331,0.002129801,0.004677087,0.05388952],"genre_scores_gemma":[0.9530715,0.0003026807,0.04070783,0.0001247602,0.00004624787,0.00008268725,0.0006841074,0.0002455153,0.004734593],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01197782,"threshold_uncertainty_score":0.04006982,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03055899192423368,"score_gpt":0.26428625476991,"score_spread":0.2337272628456763,"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."}}