{"id":"W7130543099","doi":"10.1109/fllm67465.2025.11391103","title":"Agent-guided Causal Discovery with a Small Language Model","year":2025,"lang":"","type":"article","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"Queen's University","funders":"","keywords":"Spurious relationship; Causal model; Language model; Equivalence (formal languages); Task (project management); Causal structure; Causal inference; Benchmark (surveying)","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":[0.004661603,0.001049691,0.001450455,0.00205375,0.001120843,0.001587023,0.002997703,0.001712751,0.003691657],"category_scores_gemma":[0.02643712,0.0009636029,0.001830581,0.001003838,0.002297598,0.003563025,0.003502768,0.002913375,0.0005778587],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001705751,"about_ca_system_score_gemma":0.003974774,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.008290555,"about_ca_topic_score_gemma":0.01449628,"domain_scores_codex":[0.9962728,0.002218067,0.0001494245,0.0006805126,0.0005255588,0.0001537387],"domain_scores_gemma":[0.9767297,0.01998619,0.0008318634,0.001397848,0.000695257,0.0003591811],"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":[0.0002835909,0.0002168612,0.002923062,0.0003500822,0.0002034516,0.0005296675,0.0006283767,0.7772527,0.002931601,0.1156567,0.003363372,0.09566053],"study_design_scores_gemma":[0.00002591688,0.00001383766,0.00005878429,0.000007061185,0.00001533088,0.00002156597,0.00002297657,0.9513367,0.0004485134,0.04740721,0.0006349899,0.000007088523],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.01124631,0.00009683404,0.9857693,0.000563426,0.00001769765,0.00009593535,0.0001219413,0.001212908,0.0008756144],"genre_scores_gemma":[0.3912678,0.0001142286,0.6050272,0.0004789745,0.00004869835,0.000450352,0.0005276833,0.0002696211,0.001815417],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.008290555,"threshold_uncertainty_score":0.0246532,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.04936350075707562,"score_gpt":0.2976340889346764,"score_spread":0.2482705881776008,"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."}}