{"id":"W4411337688","doi":"10.1109/cain66642.2025.00033","title":"LLM-Based Safety Case Generation for Baidu Apollo: Are We there Yet?","year":2025,"lang":"en","type":"article","venue":"","topic":"Risk and Safety Analysis","field":"Decision Sciences","cited_by":2,"is_retracted":false,"has_abstract":true,"ca_institutions":"York University","funders":"","keywords":"Apollo; Computer science; Astrobiology; Physics","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":"codex-gemma-dda1882f352a","candidate_categories":["insufficient_payload"],"consensus_categories":[],"category_scores_codex":[0.001631749,0.0001309745,0.000293837,0.0002718601,0.0004685349,0.0002167134,0.0003436223,0.0001013805,0.001326358],"category_scores_gemma":[0.0009819129,0.00008484816,0.0003124127,0.001012961,0.00004614016,0.000170359,0.00004209367,0.00006766802,0.0001291347],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00004362991,"about_ca_system_score_gemma":0.0001293109,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0001631512,"about_ca_topic_score_gemma":0.00453521,"domain_scores_codex":[0.9981122,0.0001652933,0.0005748066,0.0004759646,0.0004769669,0.0001947385],"domain_scores_gemma":[0.9974734,0.001293289,0.0001483783,0.0006599611,0.0003527883,0.00007214618],"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":[0.0002560392,0.000254309,0.005415374,0.00001841405,0.0001549209,0.0001412924,0.00025792,0.02451375,0.0008608831,0.03492678,0.4126482,0.5205521],"study_design_scores_gemma":[0.001255102,0.00007041877,0.000814326,0.00002244195,0.0001193886,0.00001104581,0.002988286,0.596997,0.003904183,0.03494546,0.3585973,0.0002750942],"study_design_candidate":"not_applicable","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.02426266,0.0007058728,0.8929536,0.05884892,0.0005991266,0.0004273563,0.0001178407,0.00007933884,0.0220053],"genre_scores_gemma":[0.9406385,0.0001225791,0.007309068,0.001604017,0.0001029447,0.00003087363,0.00001967866,0.000007452466,0.05016489],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9163758,"threshold_uncertainty_score":0.9995866,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.1572253604098817,"score_gpt":0.4148172320835699,"score_spread":0.2575918716736882,"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."}}