{"id":"W4412708207","doi":"10.1016/j.aap.2025.108178","title":"Assessing the efficacy of pre-trained large language models in analyzing autonomous vehicle field test disengagements","year":2025,"lang":"en","type":"article","venue":"Accident Analysis & Prevention","topic":"Traffic and Road Safety","field":"Engineering","cited_by":4,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"National Research Council Canada; National Highway Traffic Safety Administration; National Research Council","keywords":"Disengagement theory; Field (mathematics); Computer science; Risk analysis (engineering); Poison control; Control (management); Data science; Applied psychology; Engineering; Artificial intelligence; Psychology; Business; Medicine; Environmental health","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.005853658,0.001847672,0.0007699435,0.0009388528,0.0004252949,0.001461616,0.001431821,0.001593738,0.001799853],"category_scores_gemma":[0.02705299,0.0005720453,0.0007475303,0.0004294964,0.0004147935,0.002238713,0.001298252,0.002198074,0.00166723],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0005991077,"about_ca_system_score_gemma":0.001247014,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008800071,"about_ca_topic_score_gemma":0.01095146,"domain_scores_codex":[0.9977397,0.001001473,0.0001996668,0.0006754575,0.0002090639,0.0001745671],"domain_scores_gemma":[0.9596506,0.03492619,0.0008032955,0.001758697,0.002133383,0.0007277571],"domain_codex":null,"domain_gemma":"methods","domain_candidate":"methods","domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.007525825,0.007694114,0.2259185,0.0005705002,0.00135922,0.0004792374,0.002192703,0.1294432,0.03192949,0.0006198162,0.007856129,0.5844113],"study_design_scores_gemma":[0.000265385,0.002723915,0.07723295,0.00009717046,0.0005182007,0.0002538044,0.0008912867,0.897558,0.01820275,0.000862525,0.001256607,0.0001373969],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.9579595,0.0004493208,0.03490423,0.0002257255,0.0001866226,0.0002627412,0.001048496,0.002424971,0.002538352],"genre_scores_gemma":[0.9746245,0.0001250932,0.0206639,0.0001395773,0.00003564733,0.0001997981,0.002845991,0.0001087539,0.001256598],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.9941463,"threshold_uncertainty_score":0.03095746,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009146477039634901,"score_gpt":0.2989159307987034,"score_spread":0.2897694537590685,"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."}}