{"id":"W4318351452","doi":"10.48550/arxiv.2301.11305","title":"Cognitive Constraint Simulation and the Geometry of Human Authorship: A First-Principles Theory of AI Text Detection","year":2023,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Topic Modeling","field":"Computer Science","cited_by":154,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Canadian Institute for Advanced Research","keywords":"Discriminative model; Zero (linguistics); Computer science; Artificial intelligence; Curvature; Shot (pellet); Classifier (UML); Function (biology); Machine learning; Natural language processing; Algorithm; Pattern recognition (psychology); Mathematics; Geometry; Linguistics","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001007285,0.0001652834,0.0002970078,0.0003138671,0.0001499553,0.00003625458,0.0005953947,0.0002016594,0.000007968364],"category_scores_gemma":[0.0002929977,0.000153653,0.000135695,0.0003983747,0.0004682291,0.0001463323,0.001230884,0.0004367639,0.000003255281],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005207961,"about_ca_system_score_gemma":0.00007005769,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.000200841,"about_ca_topic_score_gemma":0.0001247751,"domain_scores_codex":[0.9986211,0.0003018533,0.0002693663,0.0005555722,0.0001007893,0.0001513044],"domain_scores_gemma":[0.9976426,0.001134761,0.0003909256,0.000548975,0.0002319426,0.00005076844],"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":[0.00005525341,0.00001928295,0.001530131,0.0001183097,0.00008454659,0.000008611873,0.001278923,0.590583,0.00002052748,0.4051247,2.800996e-7,0.001176436],"study_design_scores_gemma":[0.0005584251,0.00002966031,0.003938011,0.0001671119,0.00007142143,9.190539e-7,0.0003467502,0.7842777,0.0003594126,0.2101275,0.000002014737,0.0001209999],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.455535,0.00001828439,0.5438921,0.00002963167,0.00009920731,0.0002011396,0.000007260562,0.000051871,0.0001654864],"genre_scores_gemma":[0.9995044,0.00001762571,0.0001934006,0.00001687073,0.00002396045,0.000001042933,0.000003037938,0.000009412076,0.0002302515],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.5439694,"threshold_uncertainty_score":0.6265786,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.150166583143718,"score_gpt":0.2412236492640316,"score_spread":0.09105706612031361,"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."}}