{"id":"W4404609282","doi":"10.1109/tse.2024.3504286","title":"On Inter-Dataset Code Duplication and Data Leakage in Large Language Models","year":2024,"lang":"en","type":"article","venue":"IEEE Transactions on Software Engineering","topic":"Topic Modeling","field":"Computer Science","cited_by":10,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University; McGill University","funders":"","keywords":"Computer science; Programming language; Code (set theory); Software bug; Data mining; Software","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":[],"consensus_categories":[],"category_scores_codex":[0.0002007637,0.0001241825,0.0001000817,0.0002362657,0.00003663964,0.0001218547,0.0004662223,0.0000537107,0.000007382187],"category_scores_gemma":[0.00001228131,0.0001301905,0.00001872479,0.0002340447,0.000005444374,0.0007055975,0.00001404756,0.0002968762,0.00002151858],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005568402,"about_ca_system_score_gemma":0.00001708222,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.0000313411,"about_ca_topic_score_gemma":0.00005287107,"domain_scores_codex":[0.999038,0.00001268498,0.0001582677,0.0004799076,0.0001245755,0.0001865412],"domain_scores_gemma":[0.9989417,0.0001519657,0.00001034284,0.0008380847,0.000006612539,0.00005133848],"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":[0.000004488487,0.00006763657,0.000003953826,0.0001284681,0.00002301964,0.00005728487,0.001274819,0.926923,0.0004100258,0.004078231,0.0003186795,0.06671038],"study_design_scores_gemma":[0.0001225693,0.00001574709,0.00001334587,0.0001412936,0.000004724828,0.00001076835,0.00001298885,0.9981368,0.0006579716,0.00009312145,0.0006645594,0.0001261421],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.008487405,0.0001805354,0.9894779,0.0001056322,0.0003785571,0.0001063373,0.0007817365,0.0004759704,0.000005905858],"genre_scores_gemma":[0.9559272,0.00002803016,0.04378631,0.00007981462,0.00002253611,0.00002281495,0.00007978183,0.00001922946,0.000034282],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.9474398,"threshold_uncertainty_score":0.5309016,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02599332964216474,"score_gpt":0.2756331319817165,"score_spread":0.2496398023395518,"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."}}