{"id":"W4416750974","doi":"10.1109/iros60139.2025.11246348","title":"AGENTS-LLM: Augmentative GENeration of Challenging Traffic Scenarios with an Agentic LLM Framework","year":2025,"lang":"","type":"article","venue":"","topic":"Autonomous Vehicle Technology and Safety","field":"Engineering","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Key (lock); Control (management); Domain (mathematical analysis); Quality (philosophy); Scenario testing; Scratch; Scale (ratio)","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":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.0003108489,0.0004890754,0.0005991453,0.0004344645,0.0003470154,0.00004685955,0.0003953836,0.0006823865,0.0008700569],"category_scores_gemma":[0.00003122342,0.000487168,0.0001226851,0.0007910178,0.0002651876,0.0003607867,0.00008798507,0.0008144295,0.00003670285],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002772828,"about_ca_system_score_gemma":0.0001518768,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00002633757,"about_ca_topic_score_gemma":0.0001196624,"domain_scores_codex":[0.9976624,0.0001072772,0.0007592548,0.0006245009,0.0002515565,0.0005950587],"domain_scores_gemma":[0.9988118,0.00009382304,0.0001617871,0.0007060333,0.0001154794,0.0001110084],"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":[0.0001506461,0.001066709,0.0004239455,0.0005199806,0.00142555,0.00004141562,0.007483892,0.8490171,0.002319503,0.06237857,0.0001675851,0.07500513],"study_design_scores_gemma":[0.001086208,0.0007110404,0.00184303,0.000490353,0.0003470742,0.000006527293,0.004007251,0.9711646,0.01889435,0.0006971948,0.0002043886,0.0005479794],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.6672263,0.0009962525,0.3270344,0.0005220424,0.0004671326,0.0006095305,0.000009812553,0.0004802877,0.002654176],"genre_scores_gemma":[0.9893447,0.0006545589,0.00884151,0.0001517183,0.00008135197,0.00004834363,0.00002994015,0.00005822554,0.0007896277],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.3221184,"threshold_uncertainty_score":0.999758,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02481226748467035,"score_gpt":0.2696134597636205,"score_spread":0.2448011922789501,"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."}}