{"id":"W4416251197","doi":"10.1109/ijcnn64981.2025.11229064","title":"Bridging AI and Regulation: Large Language Models for Documentation Compliance Check","year":2025,"lang":"","type":"article","venue":"","topic":"Ethics and Social Impacts of AI","field":"Social Sciences","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"PricewaterhouseCoopers (Canada)","funders":"","keywords":"Documentation; Bridging (networking); Trustworthiness; Accountability; Transparency (behavior); Compliance (psychology); 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":[0.009511514,0.001138046,0.0007363833,0.003550105,0.001120002,0.003419589,0.001969056,0.001765558,0.003215032],"category_scores_gemma":[0.05072284,0.0004532482,0.002131159,0.002386194,0.0009861065,0.004663867,0.002581035,0.003902887,0.002054382],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002703159,"about_ca_system_score_gemma":0.003621598,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0159171,"about_ca_topic_score_gemma":0.02942753,"domain_scores_codex":[0.9905246,0.006709764,0.0006380135,0.0009696427,0.0009254697,0.0002325448],"domain_scores_gemma":[0.9487017,0.04103271,0.002587768,0.004274216,0.002744486,0.000659174],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.001637,0.001450492,0.06199842,0.001361114,0.0006665426,0.0007795416,0.002914894,0.2851503,0.004832007,0.04626201,0.0647633,0.5281844],"study_design_scores_gemma":[0.0001032899,0.00008806414,0.003596699,0.0001858323,0.00006416377,0.00009361692,0.0004147748,0.939544,0.001610708,0.04104069,0.0132069,0.00005127773],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.2198234,0.004483797,0.7143965,0.009386757,0.0005717597,0.001064132,0.01937662,0.01815936,0.01273763],"genre_scores_gemma":[0.6056576,0.0006470545,0.3595907,0.001130724,0.0001873519,0.001037658,0.02838453,0.0006728104,0.00269153],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.0159171,"threshold_uncertainty_score":0.05030233,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05594809343478166,"score_gpt":0.4281153474471013,"score_spread":0.3721672540123196,"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."}}