{"id":"W6892423477","doi":"10.5281/zenodo.10152237","title":"Whodunit: Classifying Code as Human Authored or GPT-4 Generated - A case study on CodeChef problems","year":2024,"lang":"en","type":"article","venue":"Zenodo (CERN European Organization for Nuclear Research)","topic":"Artificial Intelligence in Healthcare and Education","field":"Medicine","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Classifier (UML); Stylometry; Source code; Genetic programming; Code (set theory); Generative grammar","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":[0.001703238,0.001234272,0.0005788813,0.00405616,0.001092217,0.001541853,0.001780612,0.002176667,0.005223152],"category_scores_gemma":[0.01340648,0.0002792935,0.0009443153,0.003388659,0.0008956019,0.002201113,0.001551498,0.001740952,0.005533427],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.002772024,"about_ca_system_score_gemma":0.001291768,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.02329865,"about_ca_topic_score_gemma":0.04610602,"domain_scores_codex":[0.997296,0.0005685006,0.0002691066,0.0006943909,0.0009323053,0.0002395713],"domain_scores_gemma":[0.9914982,0.003629363,0.0006285304,0.002079262,0.001710137,0.000454476],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"observational","study_design_scores_codex":[0.0008965399,0.0008395552,0.04756118,0.00185807,0.0001149908,0.000977201,0.001290508,0.009249797,0.003281305,0.004686268,0.7744026,0.154842],"study_design_scores_gemma":[0.0005766927,0.0007344755,0.1023125,0.0007135263,0.00008395532,0.003096371,0.00238408,0.160542,0.01881398,0.008351225,0.7021739,0.0002174334],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.4550019,0.004079435,0.01746035,0.003439021,0.0008952573,0.001095899,0.4509442,0.02925025,0.03783369],"genre_scores_gemma":[0.2375298,0.0007228317,0.03670125,0.0008372557,0.0001421581,0.0007898687,0.7082422,0.001882797,0.01315191],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.02329865,"threshold_uncertainty_score":0.0463261,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.422070655404073,"score_gpt":0.4605106813435444,"score_spread":0.03844002593947143,"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."}}