{"id":"W7117566378","doi":"10.1109/rasse64831.2025.11315433","title":"Can Large Language Models Assist with SOTIF Scenario Generation?","year":2025,"lang":"","type":"article","venue":"","topic":"Safety Systems Engineering in Autonomy","field":"Engineering","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Critical Systems Labs","funders":"","keywords":"Identification (biology); Task (project management); Set (abstract data type); Natural language; Natural language generation; Modeling language; Control (management); Automotive industry","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":[0.005232593,0.001454329,0.0005680034,0.001185203,0.0005676102,0.002982218,0.002269443,0.001802278,0.006340279],"category_scores_gemma":[0.0373634,0.0009440577,0.001203986,0.000611593,0.000935311,0.006778385,0.002441085,0.002268253,0.002742115],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0009277349,"about_ca_system_score_gemma":0.001566816,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.003159129,"about_ca_topic_score_gemma":0.005359817,"domain_scores_codex":[0.9968958,0.002059198,0.0001488316,0.0004304914,0.0003623627,0.0001033518],"domain_scores_gemma":[0.9719579,0.02307569,0.0009358852,0.002477891,0.001183416,0.0003691722],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.001255524,0.001264267,0.01117606,0.001357465,0.0003073252,0.001101632,0.00704062,0.2364387,0.04533352,0.06809112,0.02462598,0.6020078],"study_design_scores_gemma":[0.0001754424,0.0001770036,0.0007279169,0.0001462124,0.00006930498,0.0002859753,0.0009464664,0.9073259,0.01200524,0.05487841,0.02316718,0.00009499137],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.048297,0.0002016355,0.928254,0.002054773,0.0001172655,0.0005735046,0.0008358755,0.01283452,0.00683146],"genre_scores_gemma":[0.3079454,0.0002263307,0.6863661,0.0005070605,0.00003906613,0.0004969189,0.001956675,0.001047571,0.001414846],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.006340279,"threshold_uncertainty_score":0.02767289,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.008582446114813725,"score_gpt":0.2106372941209732,"score_spread":0.2020548480061595,"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."}}