{"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":"metacan-v3-hybrid-931329e0061c","candidate_categories":["metaresearch"],"consensus_categories":[],"category_scores_codex":[0.02872478,0.001498598,0.002957208,0.003088194,0.003063755,0.004560186,0.005749554,0.003566684,0.002535812],"category_scores_gemma":[0.1562303,0.001882936,0.002995149,0.00584463,0.004021517,0.01361263,0.008428127,0.00443435,0.001055773],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.006384215,"about_ca_system_score_gemma":0.006746542,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01148668,"about_ca_topic_score_gemma":0.01410811,"domain_scores_codex":[0.9706745,0.01399995,0.002330345,0.006844714,0.004346288,0.001804118],"domain_scores_gemma":[0.7697986,0.1693968,0.009390376,0.04331812,0.006304268,0.001791798],"domain_codex":null,"domain_gemma":"reproducibility","domain_candidate":"reproducibility","domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"observational","study_design_scores_codex":[0.002395484,0.0009109406,0.05993858,0.0008097779,0.001004961,0.00152639,0.001972378,0.4896173,0.004468297,0.05866565,0.0373673,0.3413229],"study_design_scores_gemma":[0.0001069256,0.0001556763,0.002512796,0.00008018933,0.0001307202,0.0004471523,0.0002652523,0.9181488,0.003195175,0.0723694,0.002545991,0.00004208199],"study_design_candidate":"observational","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.3296938,0.004565615,0.6286742,0.01354997,0.0005207661,0.0006511637,0.004748337,0.01285495,0.004741189],"genre_scores_gemma":[0.833484,0.0008029927,0.1488988,0.002982575,0.0003831496,0.0006379576,0.006726351,0.001500207,0.004584006],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.9712752,"threshold_uncertainty_score":0.1519129,"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."}}