{"id":"W7130549209","doi":"10.1109/fllm67465.2025.11390932","title":"Explainable AI via Large Language Models: Translating Neural Network Behavior into Interpretable Decision Trees","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":"Mila - Quebec Artificial Intelligence Institute; Artificial Intelligence in Medicine (Canada); Université TÉLUQ","funders":"","keywords":"Artificial neural network; Decision tree; Fidelity; Deep neural networks; Audit; Deep learning; Language model; Language understanding","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","sts","scholarly_communication"],"consensus_categories":[],"category_scores_codex":[0.002163421,0.0009975961,0.001086327,0.0006390861,0.001873021,0.001933042,0.003438806,0.0005322348,0.0008852165],"category_scores_gemma":[0.000189049,0.001008597,0.0005765227,0.003721666,0.000224248,0.005641191,0.001817269,0.001180598,0.000267484],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0004251238,"about_ca_system_score_gemma":0.0004444491,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.004151366,"about_ca_topic_score_gemma":0.01054558,"domain_scores_codex":[0.9915173,0.0005093507,0.002049555,0.002159348,0.0009739088,0.00279051],"domain_scores_gemma":[0.9956285,0.0007272905,0.0003189437,0.00233764,0.0005255679,0.0004620386],"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":[0.0001953355,0.0007741706,0.0009336804,0.0001212574,0.00008512397,0.0003280742,0.02277741,0.2197345,0.001939507,0.0868404,0.002941658,0.6633289],"study_design_scores_gemma":[0.0004782224,0.0002918737,0.00006230286,0.0006264674,0.0001291962,0.00001877864,0.004562431,0.9281839,0.007701021,0.05580566,0.001260075,0.0008800465],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.03987705,0.006412903,0.933668,0.001586054,0.003079139,0.001357745,0.000007637952,0.000536755,0.01347475],"genre_scores_gemma":[0.9325958,0.0001000297,0.05775811,0.002658135,0.0002826519,0.0002290567,0.00001076046,0.0000714755,0.006293988],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8927187,"threshold_uncertainty_score":0.9994264,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01935430610804568,"score_gpt":0.3086205793526522,"score_spread":0.2892662732446065,"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."}}