{"id":"W7126202239","doi":"10.18280/isi.301203","title":"Interpretable Multi-Label Classification of Human- and Large Language Model-Generated Texts Using Transformer Embeddings and Explainable Artificial Intelligence","year":2025,"lang":"","type":"article","venue":"Ingénierie des systèmes d information","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":false,"ca_institutions":"","funders":"","keywords":"Transformer; Natural language; Pattern recognition (psychology); Applications of artificial intelligence; Language model","routes":{"ca_aff":false,"ca_fund":false,"ca_venue":true,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.002003025,0.001034327,0.0006106395,0.001536791,0.0004963533,0.00193173,0.001118922,0.001458203,0.002381443],"category_scores_gemma":[0.01117797,0.0003338894,0.001113461,0.0007542198,0.0008092063,0.003444228,0.001649502,0.002145419,0.001021262],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001010137,"about_ca_system_score_gemma":0.0007026644,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.002208626,"about_ca_topic_score_gemma":0.003538867,"domain_scores_codex":[0.9983042,0.0007774766,0.0001091217,0.0004619048,0.0002167633,0.0001305067],"domain_scores_gemma":[0.9893368,0.007659052,0.0005739719,0.001125003,0.001105835,0.0001992965],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.002403327,0.00101938,0.01616327,0.0007057605,0.0003897823,0.001658124,0.002448533,0.2571781,0.03096571,0.04286765,0.01886163,0.6253387],"study_design_scores_gemma":[0.00002433224,0.00004815964,0.0008026378,0.00002147118,0.0000262943,0.00005419213,0.0001184986,0.9733557,0.003828505,0.02094232,0.0007642568,0.0000136491],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2010348,0.0004833487,0.7885447,0.001285653,0.0003043456,0.0001393444,0.001809298,0.004083759,0.002314881],"genre_scores_gemma":[0.8540799,0.0001249648,0.1384259,0.0001572344,0.0001271033,0.0001177023,0.00427948,0.0003378653,0.002349999],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.002381443,"threshold_uncertainty_score":0.01059318,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0527775078637612,"score_gpt":0.3249960898983826,"score_spread":0.2722185820346215,"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."}}